2017 Book InformaticsInSchoolsFocusOnLea

Valentina Dagiene
Arto Hellas (Eds.)

LNCS 10696
Informatics in Schools
Focus on Learning Programming
10th International Conference on Informatics in Schools:
Situation, Evolution, and Perspectives, ISSEP 2017
Helsinki, Finland, November 13–15, 2017, Proceedings
123
Lecture Notes in Computer Science 10696
Commenced Publication in 1973
Founding and Former Series Editors:
Gerhard Goos, Juris Hartmanis, and Jan van Leeuwen
Editorial Board
David Hutchison
Lancaster University, Lancaster, UK
Takeo Kanade
Carnegie Mellon University, Pittsburgh, PA, USA
Josef Kittler
University of Surrey, Guildford, UK
Jon M. Kleinberg
Cornell University, Ithaca, NY, USA
Friedemann Mattern
ETH Zurich, Zurich, Switzerland
John C. Mitchell
Stanford University, Stanford, CA, USA
Moni Naor
Weizmann Institute of Science, Rehovot, Israel
C. Pandu Rangan
Indian Institute of Technology, Madras, India
Bernhard Steffen
TU Dortmund University, Dortmund, Germany
Demetri Terzopoulos
University of California, Los Angeles, CA, USA
Doug Tygar
University of California, Berkeley, CA, USA
Gerhard Weikum
Max Planck Institute for Informatics, Saarbrücken, Germany
More information about this series at http://www.springer.com/series/7407
Valentina Dagiene Arto Hellas (Eds.)
•
Informatics in Schools
Focus on Learning Programming
10th International Conference on Informatics in Schools:
Situation, Evolution, and Perspectives, ISSEP 2017
Helsinki, Finland, November 13–15, 2017
Proceedings
123
Editors
Valentina Dagiene Arto Hellas
Mathematics and Informatics University of Helsinki
Vilnius University Helsinki
Vilnius Finland
Lithuania
ISSN 0302-9743 ISSN 1611-3349 (electronic)

Lecture Notes in Computer Science
ISBN 978-3-319-71482-0 ISBN 978-3-319-71483-7 (eBook)
https://doi.org/10.1007/978-3-319-71483-7
Library of Congress Control Number: 2017960853
LNCS Sublibrary: SL1 – Theoretical Computer Science and General Issues
© Springer International Publishing AG 2017

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the
material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation,
broadcasting, reproduction on microfilms or in any other physical way, and transmission or information
storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now
known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this book are
believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors
give a warranty, express or implied, with respect to the material contained herein or for any errors or
omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in
published maps and institutional affiliations.
Printed on acid-free paper
This Springer imprint is published by Springer Nature

The registered company is Springer International Publishing AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Preface
The International Conference on Informatics in Schools: Situation, Evolution and

Perspectives (ISSEP) celebrated its 10th anniversary in Helsinki, the capital of Finland.
The year 2017 also marked the anniversary of Finland: One month after the conference,
on December 6, Finland celebrated its 100th year of independence.
ISSEP is a forum for researchers and practitioners in the area of informatics edu-
cation, in both primary and secondary schools. The conference provides an opportunity
for educators and researchers to reflect upon the goals and objectives of this subject
matter, its curricula, various teaching and learning paradigms and topics, as well as the
connections to everyday life – including the various ways of developing informatics
education in schools. This applies in an even larger extent to the didactics of infor-
matics. ISSEP provides an opportunity to take stock of developments in the field and
on the extent these developments are relevant for everyday school life.
This conference brings together delegates to address pressing issues in informatics
education. On the basis of these challenging areas, participants consider how to cope
with future demands of education in informatics and IT-related subjects. Teachers,
those working in teacher education and in-service training as well as in school
administration come to participate in workshops and to share their experience or vision
on an international platform.
The ISSEP conference started in 2005 in Austria, Klagenfurt University (now
Alpen-Adria University of Klagenfurt), where the conference was a part of the cele-
bration of 20 years of teaching informatics in Austrian secondary schools. The fol-
lowing year, ISSEP took place in Vilnius, Lithuania, celebrating 20 years of
informatics in Lithuanian secondary schools. The ISSEP conference expanded: In
Vilnius there were over 200 researchers from 37 countries, and an additional day for
local teachers was established and had over 250 attendants from all over Lithuania. In
2008, ISSEP was in Torun, Poland, and in 2010, ISSEP was organized in Zürich. Then
followed Bratislava, Slovakia, in 2011, Oldenburg, Germany, in 2013, Istanbul,
Turkey, in 2014, Ljubljana, Slovenia, in 2015 and Münster, Germany, in 2016.
In these proceedings, all research papers were blind reviewed by at least three
reviewers. A final selection phase was conducted by the co-chairs who reviewed all
reviews and recommendations before making their final decisions. From the 41 sub-
mitted articles, 18 were selected for presentation and inclusion in the proceedings.
This year, participants at ISSEP discussed a broad range of themes ranging from
making informatics accessible to visually impaired students and computational think-
ing to context- and country-specific challenges as well as teacher development and
training. The conference hosted several workshops where teachers and researchers
were familiarized with, e.g., robotics and teaching programming as well as unplugged
activities for computational thinking. Preliminary work and new ideas were presented
VI Preface
as posters, creating a unique opportunity for the participants to discuss the ideas in a
supportive environment. ISSEP also hosted CSERC (Computer Science Education
Research Conference) as a special track at the conference.
October 2017 Valentina Dagienė

Arto Hellas
Organization
Conference Chairs and Co-chairs

Valentina Dagienė Vilnius University, Lithuania
Arto Hellas University of Helsinki, Finland
Program Committee
Erik Barendsen Radboud University Nijmegen and Open Universiteit,
The Netherlands
Andreas Bollin University of Klagenfurt, Austria
Andrej Brodnik University of Ljubljana, Slovenia
Michalis Giannakos Norwegian University of Science and Technology,
Norway
David Ginat Tel Aviv University, Israel
Bruria Haberman Holon Institute of Technology, Tel Aviv, Israel
Juraj Hromkovič Swiss Federal Institute of Technology Zurich,
Switzerland
Peter Hubwieser Technical University of Munich, Germany
Petri Ihantola Tampere University of Technology, Finland
Ivan Kalas Comenius University, Slovakia
Tiina Korhonen University of Helsinki, Finland
Peter Micheuz University of Klagenfurt and Gymnasium Völkermarkt,
Austria
George A. Papadopoulos University of Cyprus, Cyprus
Sergey Pozdniakov Saint Petersburg Electrotechnical University, Russia
Ralf Romeike University of Erlangen (FAU), Germany
Maciej M. Syslo University of Wroclaw, Poland
Local Organizing Committee

Antti Laaksonen Aalto University, Finland
Juho Leinonen University of Helsinki, Finland
Leo Leppänen University of Helsinki, Finland
Contents
Keynote Paper
The Computer Science Way of Thinking in Human History

and Consequences for the Design of Computer Science Curricula . . . . . . . . . 3
Juraj Hromkovič and Regula Lacher
Accessibility, Visualizations and Physical Computing
Adolescent and Adult Student Attitudes Towards Progress Visualizations. . . . 15

Onni Aarne, Petrus Peltola, Antti Leinonen, Juho Leinonen,
and Arto Hellas
3D Printing as Medium for Motivation and Creativity in Computer

Science Lessons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Petra Kastl, Oliver Krisch, and Ralf Romeike
Teaching Materials in Informatics for Lower Secondary

School Blind Students . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
Ľudmila Jašková and Mária Stankovičová
Teachers’ Expectations and Experience in Physical Computing . . . . . . . . . . . 49

Mareen Przybylla, Finn Henning, Carla Schreiber,
and Ralf Romeike
Teacher Development and Assessment
Investigating Informatics Teachers’ Initial Pedagogical Content

Knowledge on Modeling and Simulation . . . . . . . . . . . . . . . . . . . . . . . . . . 65
Nataša Grgurina, Erik Barendsen, Cor Suhre, Klaas van Veen,
and Bert Zwaneveld
A Notation for Sets, Sequences and Series: Possible Benefits

for Understanding and Use . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
Susanna H. du Plessis and Vreda Pieterse
Bebras Tasks
Bebras Task Analysis in Category Little Beavers in Slovakia . . . . . . . . . . . . 91

Lucia Budinská, Karolína Mayerová, and Michaela Veselovská
X Contents
Promoting Computational Thinking Skills: Would You Use

this Bebras Task? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
Annalisa Calcagni, Violetta Lonati, Dario Malchiodi, Mattia Monga,
and Anna Morpurgo
Country Reports
Introducing Programming and Digital Competence

in Swedish K-9 Education . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
Fredrik Heintz, Linda Mannila, Lars-Åke Nordén, Peter Parnes,
and Björn Regnell
Preparedness of Japan’s Elementary School Teachers for the Introduction

of Computer Programming Education . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
Yutaro Ohashi
Baltic and Nordic K-12 Teacher Perspectives on Computational

Thinking and Computing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
Arnold Pears, Valentina Dagiene, and Egle Jasute
Computational Thinking/Programming, Computational Abilities
Identifying Students’ Misconceptions on Basic Algorithmic Concepts

Through Flowchart Analysis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
Ebrahim Rahimi, Erik Barendsen, and Ineke Henze
A Preliminary Investigation on Computational Abilities

in Secondary School . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
Ugo Solitro, Margherita Pasini, Debora De Gradi,
and Margherita Brondino
Solving Everyday Challenges in a Computational Way of Thinking . . . . . . . 180

Bernhard Standl
System Papers
Teaching Basic Elements of OOP in School Informatics During

Constructing Virtual Micro-worlds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195
Evgeny A. Eremin
Real-Time Data Analyses in Secondary Schools Using a Block-Based

Programming Language . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207
Andreas Grillenberger and Ralf Romeike
XLogoOnline: A Single-Page, Browser-Based Programming

Environment for Schools Aiming at Reducing Cognitive Load on Pupils . . . . 219
Juraj Hromkovič, Giovanni Serafini, and Jacqueline Staub
Contents XI
Introduction to Bebras Challenge Management: Overview

and Analyses of Developed Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232
Valentina Dagiene, Gabriele Stupuriene, Lina Vinikiene,
and Rimantas Zakauskas
Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245

Keynote Paper
The Computer Science Way of Thinking
in Human History and Consequences
for the Design of Computer Science Curricula
Juraj Hromkovič(B) and Regula Lacher
Department of Computer Science, ETH Zurich, Zurich, Switzerland

juraj.hromkovic@inf.ethz.ch
Abstract. Teaching computer science offers more than algorithmic

thinking (or more general and as recently presented: computational
thinking). To understand this claim, one has to have a more careful look
at the development of human culture, science, and technology. This helps
not only to recognize that the computer science way of thinking was cru-
cial for the development of human society since anyone can remember,
but it helps to make a good choice of topics for sustainable computer
science education in the context of science and humanities. This leads to
the creation of textbooks that do not focus on particular knowledge for
specialists, but offer serious contributions in the very general framework
of education.
1 Introduction
The recent technological development has led to a situation where many of our
children will be working in a profession that does not yet exist, and they will
be working on the solution of problems that have not even materialized yet.
To prepare them for this, we need to change our approach to teaching. Our
fundamental principle for designing educational curricula for any subject is the
following:
Do not teach the final products of science, technology, and humanities, and
do not consider it the highest goal to train to successfully apply them. For
the latest knowledge may be found outdated with time. Teach the process
of discovering new knowledge, teach the need to search for new solutions,
teach the ways of collecting experience and formulating hypotheses, teach
the ways of verifying hypotheses, teach how others can be convinced about
the truth discovered, teach the constructive way of thinking in order to
create new products and finally new technology, and teach the processes of
testing and improving the products of our work.
We consider this principle to be the base of the so-called “critical thinking,”

becoming the fundamental goal of teaching science in leading universities. It is
based on the understanding of the development of human societies, which was
c Springer International Publishing AG 2017
V. Dagiene and A. Hellas (Eds.): ISSEP 2017, LNCS 10696, pp. 3–11, 2017.
https://doi.org/10.1007/978-3-319-71483-7_1
4 J. Hromkovič and R. Lacher
only possible with the evolvement of science, humanities, law, and technology.
We should teach those aspects that have the biggest impact on the development
of our way of thinking. The whole of human history can be seen as a process of
developing research instruments to understand the world around us and to find
solutions for our problems. The goal of this article is to look at computer science
as an integral part of science since ever and to use this view to recognize the
main contributions and consequently to design how to teach them in the context
of other educational subjects.
2 History
2.1 Computer Science, Languages, and Writing
Writing was developed in many cultures, and it became one of the most
important human technologies. The oldest known writing was developed in
Mesopotamia more than 5 000 years ago. The motivation for this development
would be recently considered a “pure computer science task”: The people of
Mesopotamia needed to save (store) and process (update) tax and property
data of some 1 000 000 people living between Euphrates and Tigris. The start-
ing point during the development of a writing is the choice of symbols, called
an alphabet, and then the representation of data (numbers, words, texts) as
finite sequences of symbols. In this framework, we can speak about “coding” of
information. The development of number representations is a wonderful story
about different systems being in competition with each other (each to each) for
many years with respect to understandability (transparency), the length of the
description, and the efficiency of performing arithmetic operations on them.
As more and more people learned to read, the need to keep some of that writ-
ten information secret, i.e., understandable for a selected group of people only,
was formulated. The oldest traces of attempts to decrease readability are some
3 500 years old, where the order of some symbols was exchanged in Mesopotamia.
2 500 years ago, different secret writings were used in Palestine, India, China,
Egypt and Greece. They used concepts of mixing the order of symbols and
exchanging symbols (coding of symbols by other symbols) which are still used
in modern cryptographic protocols.
The above mentioned activities are all strongly related to the expertise of
computer scientists, some of them are even considered as a proper part of com-
puter science, e.g., cryptography. Indeed, computer scientists are true experts in
developing writings for specific purposes. Some examples for teaching this topic
can be found in [1,5,8].
Another example is compression, which is strongly related to measuring the
information content of texts. Coding texts as short as possible, without losing
any information, in order to shorten communication messages, was developed
even before computer science was established as an independent discipline (see
[1,8] for an example of a teaching sequence).
Self-correcting codes [10] are robust writings which automatically recognize
and correct mistakes in texts. Current e-commerce cannot work without such
The Computer Science Way of Thinking in Human History 5
codes because any small misprint could cause a wrong money transfer or wrong
online order.
Later on, computer scientists developed numerous writings for data rep-
resentations for different purposes. Some focused on performing selected data
operations (data structures), others focused on creating databases (information
systems) with a very fast search capability (see [1,8]).
Driven by the construction of (physical) computers, the binary alphabet and
many ways of representing data by sequences of bits were introduced1 (see [1,8]
for a teaching sequence).
Another strong relationship between languages and computer science is in the
development of programming languages. Programming languages are languages
with very well defined syntax and semantics, and looking at the development of
formal language theory as a product of computer science one observes a big inter-
section with the development of linguistics. The formal concept of a grammar,
context-sensitivity and context-freeness are all concepts developed in coopera-
tion with linguistics (see [2] for a textbook for high schools on this topic).
2.2 Computer Science and Mathematics
If one wants to understand what mathematics is about, it is helpful to view

mathematics as a language with the following two features:
(i) All sentences have an unambiguous meaning for everybody mastering the
language of mathematics.
(ii) All argumentation in mathematics is verifiable.
The demand to create a language such as mathematics was obvious. There is no

objectivity in science if one cannot communicate in such a way that everybody
interprets each claim in the same way. Additionally, without verifiable argumen-
tation the notion of truth is very relative (see [11] for more details).
This is how mathematics, together with experiments, became the main
research instrument of humans. The relation with computer science lies mainly
in abstraction and in the development of algorithms as a method for solving
different tasks.
Computer science uses digital technology and so it cannot omit the formal
language of mathematics that codes everything as sequences of symbols of a
fixed alphabet. Abstraction in computer science means to be able to represent
objects and real situations by mathematical concepts and finally as sequences
of bits. All this is a part of the expertise of computer scientists as well. The
contribution of computer science is using the formal language of mathematics to
describe computation processes and to investigate them.
1
It was technically much easier to build processors executing operations using binary
numbers than using decimal numbers. Also, it was much easier to build stable phys-
ical systems for storing information with two states only (interpreted as 0 and 1)
than developing a computer memory based on many states.
Algorithm is a key word of computer science, but this term is not new.
Already Euclid formulated algorithms in his “Elements” [4], among them the
famous Euclid’s algorithm. The term “algorithm” is due to al-Khwarizmi, who
wrote a book about Indian digits around the year 825. Humans tried to develop
algorithms as long as anyone can remember. People generated knowledge in order
to understand the world around them and especially to apply this knowledge to
develop “procedures” to reach their goals. This was key for the development of
the first human societies because it allowed jumps in performance and efficiency
increase. The point is that becoming an expert in performing a previously devel-
oped procedure was much easier than to learn to develop such procedures or
to discover new facts. One nice example from ancient Greece is the Theorem of
Pythagoras that became a procedure applied in the construction of buildings.
We know due to Pythagoras that 32 + 42 = 52 and that the triangle with sides of
length 3, 4, and 5 units always contains a right angle. Therefore, to create a right
angle, it is sufficient to take 3 ropes of the corresponding lengths (3, 4, and 5
units) to construct the right angle. There is a huge gap between the qualification
of creating this triangle and the intellectual potential to discover the Theorem
of Pythagoras, not to mention the capability to prove why it is true. In this way,
many technologies became available to big parts of society in spite of the fact
that only few people had the expertise to understand them or even to be able
to develop them.
The above was also the reason why Leibniz described mathematics as a sci-
ence of automation of human work. The idea of Leibniz was to translate the
problems of the real world into the language of mathematics and then to solve
them by formal calculations. His first dream was to develop a calculus for logical
argumentation similar to formal arithmetics. This dream was satisfied 200 years
later when logic was developed and became the formal system for automatic cal-
culation of the correctness of mathematical proofs. The second dream of Leibniz
was to create a type of mathematics in which all real problems can be described
and solved.
During the time of technical revolution, Hilbert [7] specified Leibnizs second
dream more precisely. He considered a problem as a collection of potentially
infinitely many problem instances. An algorithm for solving the problem was
therefore a method that was able to calculate the correct solution for any of the
instances of the problem. The dream of Hilbert included finding an algorithm for
each problem that can be formulated in the language of mathematics. However,
in 1930, Kurt Gödel [6] proved that there exist claims in mathematics for which
the truth cannot be decided inside of mathematics. In other words, Kurt Gödel
proved that the description power of the language of mathematics is stronger
than its argumentation power. One can formulate claims in mathematics for
which there exist no proofs whether they are correct or not. This finding was the
nucleus that started the founding of “computability” as the first sub-discipline of
computer science with the goal to classify problems into algorithmically solvable
and algorithmically unsolvable ones. The related formal definitions of the term
“algorithm” offered by Turing [18] and Church [3] are now considered as the
birth of computer science.
There are so many intersections between mathematics and computer science

in the subsequent development, that we do not try to list them here.
2.3 When and Why Computer Science Became an Independent

Discipline
Computer science became a distinguished discipline when the following two con-
ditions were satisfied:
1. One was able to develop algorithms that were unambiguously described in
the sense that no human intellect (no expert knowledge, no interpretation)
was needed to execute them.
2. (Digital) technologies were available that enabled the delegation of execution
(of precisely described procedures) to machines.
Hence, the two main components of computer science are:
1. To use mathematics to develop – in cooperation with specialists in the respec-
tive areas of science and practice – algorithms to automate more and more
human activities. Many mathematicians consider this (or big parts thereof)
a part of mathematics.
2. To contribute to the development of hardware and software technologies,
allowing the automation of more and more complex tasks and increasing the
performance of computing technologies. A part of this job is the development
of programming languages enabling humans to communicate with machines
to control them, and to instruct them to execute different activities. This
component of computer science is assigned to engineering.
There is no strict boundary between these two and many tasks require the
expertise of both components.
3 How to Teach Computer Science

One of the biggest mistakes in the past was to teach the use (application) of dig-
ital technologies (e.g., ECDL) instead of the foundations of computer science. If
the focus is set on teaching the latest developments in computer science concepts
and digital technology, this mistake is continued. Computer science has to learn
from physics which typically is taught by following the historical development of
our understanding of the physical world. Trying to teach quantum mechanics or
relativity theory is of little value for most pupils and tends to end in frustration
for most. One has to start with Galileo Galilei and Newton in order to get a first
intuition about how the physical world is functioning and in order to learn how
to design experiments that enable to verify hypotheses or disprove them. One
has to follow the genesis of science, learn from mistakes and especially learn to
develop research instruments, namely experiments and mathematics.
If we want to enable future generations to contribute to science and technol-
ogy, we have to follow the development of basic concepts of computer science
step by step. In this way we can contribute to
(i) better understanding mathematics and languages and their development,

(ii) understanding computer science as a research instrument in science and
humanities, and
(iii) understanding our technical world and being able to control and develop it.
Rather than dedicating the space of this article to explaining in detail the
spiral curriculum of teaching computer science developed at the Center for Com-
puter Science Education of ETH, we give some selected examples to illustrate
the approach we followed during the development of our textbooks.
3.1 Data Representation

The first goal is to learn that there is freedom of choice regarding the alphabet
used to code information. One can start with different number representations
and compare them with respect to transparency, length of the description, and
efficiency in working with them. One learns to recognize what they have in
common and what are the differences. Finally, one does not only learn the binary
representation of numbers and its advantages and disadvantages, but also to
design new number representations (see [1,8] for teaching examples).
Data compression is another topic that is best taught following the historical
development. One is not allowed to push to learn the best compression method
currently known. The focus must be on learning from examples in order to be able
to propose own compression methods and to compare them with other methods
presented. One can distinguish between compression methods without and with
loss of information. In the latter case, one can play with the tradeoff between the
size of the data representation achieved by a compression on one hand, and the
amount of information lost on the other hand (see [1,8] for teaching examples).
Teaching self-verifying codes is very well motivated by the use of digital
technologies. There are plenty of codes that can be presented as examples and
there is a lot of freedom to design own codes with some required properties.
Starting with suitable examples, the main idea is designing efficiently computable
and short self-verifying codes. This can be successfully explained without any
sophisticated mathematics. Again, the main goal is to learn to design own codes
with required properties for small sets of data.
Cryptography as a theory of secret writings is probably the most gratifying
topic of data representation. One is not allowed to focus on the correct usage of
current cryptographic protocols based on deep knowledge of algebra and number
theory. Again, one has to follow the historical development step by step. One
should present historic examples of secret writings and cryptosystems, let the
pupils find their weaknesses and use these to break them. Following the devel-
opment of cryptosystems, it becomes natural to understand the concept of the
security of a cryptosystem up to the current public-key cryptography. Again,
the main achievement is the ability to design new cryptosystems and to break
them, and so to discover a demand to design better and better cryptosystems
(see [1,5,8] for teaching examples).
There are many other data representation topics. An important one deals
with suitable data representation for building databases or for special algorithmic
tasks. A key point is that there are many levels of deepness and one can create a
challenging spiral curriculum for this topic with many creative and constructive
tasks.
3.2 Algorithm Design

Algorithms as methods that solve all infinitely many instances of a problem are
a very abstract concept and should not be introduced too early. For pupils up
to class 4, one is allowed to find solutions to particular problem instances and
to develop general strategies on a very intuitive level only.
But even for older pupils, for classes 7 to 9, we do not recommend to teach
specialized, concrete algorithms that do not work if one changes the specification
of the problem just a little bit. The idea is to teach robust strategies that suc-
cessfully work for many problems. Our teaching approach starts with training
to recognize feasible solutions for a new problem instance, to represent it as a
sequence of symbols and to be able to list and count all feasible solutions for
concrete instances. Then one can introduce criteria to judge the quality of the
solutions in order to choose an appropriate one or even the best feasible solution.
This imposes a lot of understanding of the problem setting and introduces the
trouble with an exponential explosion of the number of solutions. The next step
is to teach to make compromises when looking for a good solution in situations
where one cannot list and compare all solutions because there are too many.
Again, simple robust strategies as greedy or local algorithms can be explained
and discussed. It is interesting to discover that, for some problems, they compute
optimal solutions, and for other problems they may sometimes fail by offering
solutions that are far from optimal. There is a lot of creative work in discovering
problem instances for which certain strategies behave poorly.
Again, this topic is excellent for developing a spiral curriculum starting with
collecting experience by solving simple, particular problem instances, and finish-
ing by developing clever algorithms for concrete problems in high school.
3.3 Programming
For a more detailed presentation of this topic, see [9,12–17]. The starting point
is not to teach a programming language, but rather to combine problem solving
with the communication of the designed algorithm to the computer. Start with
as few instructions as possible and force the pupils through modularity to intro-
duce new instructions and to “teach” the computer to understand them. This
way, on one hand, the pupils learn how to develop a language in order to make
the communication more efficient, and on the other hand they learn modular
design, which is one of the most fundamental concepts of engineering. Simulta-
neously, they follow the genesis of the development of programming languages to
some extent and therefore naturally do not see programming languages as final
products that must be taken as they are.
The main focus is on the following contributions:

1. To learn problem solving by building up experience with solving particular
problem instances and finally developing solving strategies.
2. To learn to describe the strategies (algorithms) developed first in a meta
language and then in exact programming language (learning fundamental
concepts of programming).
3. To discover the demand of introducing new words to a language in order to
simplify communication and to increase the understandability.
4. To learn the modular design method as a fundamental concept of engineering
and to be able to use it in various situations.
5. To learn to verify by reasoning and to test by experiments the products
(programs) of own making.
4 Conclusion
To prepare our children for the future, we need to change our approach to teach-
ing fundamentally: Rather than teaching the current products of science, tech-
nology and humanities, we should teach the process of making discoveries. We
should teach collecting experience and data, formulating and verifying hypothe-
ses, how to convince ourselves and others about the truth discovered, how to
think in a constructive way that leads to the creation of new products and new
technology, and we should teach the processes of testing and improving the prod-
ucts of our work. The example of Pythagoras’ Theorem illustrates not only how
a product of science has been applied in processes, but also shows the difference
of skills needed for the discovery of new knowledge and the application thereof.
When applying this strategy for teaching computer science, one obtains the
following principles:
1. Do not teach computer science as an isolated subject, but teach computer
science as a part of science and technology offering a deep contextual view.
Take care on contributing to knowledge transfer to other disciplines. Build
and use bridges to the development of languages and mathematics.
2. Do not teach the latest products of IT and the latest scientific discover-
ies. Follow the genesis of fundamental concepts and improve them step by
step. Create the need for these concepts and their improvements and discover
the new ideas by doing experiments, proposing and verifying solutions and
evaluating their feasibility and quality. Try to discover, for all age groups,
to which extent which concepts are feasible and how to apply learning by
doing to master them. Computer science topics such as data representation,
programming, and algorithm design have many levels of deepness, therefore
lending themselves perfectly for a spiral curriculum.
3. Teach to view IT as an enabling technology. Teach to control computers by
programming and to automate well understood activities in order to make
society more efficient. Teach to work in a constructive way, to test the func-
tionality of own products, and the modular design approach. Simply teach
the way of thinking and working of technical disciplines (engineering).
References
1. Bell, T., Witten, I.H., Fellows, M.: Computer Science Unplugged. Creative Com-
mons Attribution 2.0 (2015)
2. Böckenhauer, H.-J., Hromkovič, J.: Formale Sprachen: Endliche Automaten, Gram-
matiken, Lexikalische und syntaktische Analyse. Springer, Wiesbaden (2013).
https://doi.org/10.1007/978-3-658-00725-6
3. Church, A.: An unsolvable problem of elementary number theory. Am. J. Math.
58(2), 345–363 (1936)
4. Euclid: Elements, books I-XIII
5. Freiermuth, K., Hromkovič, J., Steffen, B., Keller, L.: Einführung in die Kryp-
tologie: Lehrbuch für Unterricht und Selbststudium. Springer, Wiesbaden (2014).
https://doi.org/10.1007/978-3-8348-2269-7
6. Gödel, K.: Über formal unentscheidbare Sätze der Principia Mathematica und ver-
wandter Systeme I. Monatshefte für Mathematik und Physik, pp. 173–198 (1931)
7. Hilbert, D.: Mathematische Probleme. Nachrichten von der königl. Gesellschaft
der Wissenschaften zu Göttingen. Mathematisch-Physikalische Klasse, pp. 253–
297 (1900)
8. Hromkovič, J.: Einfach Informatik. Datendarstellung. Klett (2018, to appear)
9. Hromkovič, J., Kohn, T.: Einfach Informatik. Programmieren. Klett (2018, to
appear)
10. Hromkovič, J., Keller, L., Komm, D., Serafini, G., Steffen, B.: Entdeckendes lernen
am beispiel fehlerkorrigierender codes. Login 168, 50–55 (2011)
11. Hromkovič, J.: Homo Informaticus: why computer science fundamentals are an
unavoidable part of human culture and how to teach them. Bull. EATCS 115,
112–122 (2015)
12. Hromkovič, J., Kohn, T., Komm, D., Serafini, G.: Combining the power of python
with the simplicity of logo for a sustainable computer science education. In: Brod-
nik, A., Tort, F. (eds.) ISSEP 2016. LNCS, vol. 9973, pp. 155–166. Springer, Cham
(2016). https://doi.org/10.1007/978-3-319-46747-4 13
13. Hromkovič, J., Kohn, T., Komm, D., Serafini, G.: Examples of algorithmic thinking
in programming education. Olympiads Inform. 10, 111–124 (2016)
14. Hromkovič, J., Kohn, T., Komm, D., Serafini, G.: Algorithmic thinking from the
start. In: Bulletin of the EATCS 121, The Education Column (2017)
15. Hromkovič, J., Serafini, G., Staub, J.: XLogoOnline: a single-page, browser-based
programming environment for schools aiming at reducing cognitive load on pupils.
In: Dagiene, V., Hellas, A. (eds.) ISSEP 2017. LNCS, vol. 10696, pp. 219–231.
Springer, Cham (2017). https://doi.org/10.1007/978-3-319-71483-7 18
16. Hromkovič, J., Steffen, B.: Why teaching informatics in schools is as important as
teaching mathematics and natural sciences. In: Kalaš, I., Mittermeir, R.T. (eds.)
ISSEP 2011. LNCS, vol. 7013, pp. 21–30. Springer, Heidelberg (2011). https://doi.
org/10.1007/978-3-642-24722-4 3
17. Serafini, G.: Programmierungerricht für Kinder und deren Lehrpersonen: Unter-
richtsmaterialien, didaktische Herausforderungen und konkrete Erfahrungen, GI-
Edition. Lecture Notes in Informatics, vol. 249, pp. 267–272 (2015)
18. Turing, A.: On computable numbers, with an application to the Entscheidung-
sproblem. Proc. London Math. Soc. 42(2), 230–265 (1936)
Accessibility, Visualizations and Physical
Computing
Adolescent and Adult Student Attitudes
Towards Progress Visualizations
Onni Aarne(B) , Petrus Peltola, Antti Leinonen, Juho Leinonen,

and Arto Hellas
University of Helsinki, Helsinki, Finland

{onni.aarne,petrus.peltola,antti.leinonen,
juho.leinonen,arto.hellas}@helsinki.fi
Abstract. Keeping students motivated for the duration of a course

is easier said than done. Contextualizing student efforts with learning
progress visualizations can help maintain engagement. However, progress
can be visualized in many different ways. So far very little research has
been done into which types of visualizations are most effective, and how
different contexts affect the effectiveness of visualizations. We compare
the effects of two different progress visualizations in an introductory pro-
gramming course. Preliminary results show that older students prefer a
visualization that emphasizes long-term progress, whereas younger stu-
dents prefer a visualization that highlights progress within a single week.
Additionally, students perform better and are more motivated when their
visualization matches their age group’s preferred visualization. Possible
explanations and implications are discussed.
Keywords: Progress visualization · Student engagement · Growth

mindset · Motivation
1 Introduction
The effort that we expect from our students can be overwhelming to the extent
that students temporarily lose track of their long-term goals and original moti-
vations – seeing the forest from the trees can be hard. Progress visualizations
can be a valuable tool for showing students the metaphorical forest, and helping
them focus on the work that they are doing. Certain visualizations can improve
students’ performance in a measured task; for example, a progress bar that mea-
sures a specific activity can increase students’ engagement with that activity [8].
These types of motivational tools can be important in labour-intensive courses,
such as introductory programming courses, and can provide additional incentives
for the students to complete the tasks they are given.
Learning progress visualization tools intended for students themselves have
mostly been researched as part of broader attempts at gamifying education.
However, this research rarely looks at specific elements of gamification with
O. Aarne and P. Peltola contributed equally to this work.
https://doi.org/10.1007/978-3-319-71483-7_2
16 O. Aarne et al.
any rigor. In this work, we compare two types of progress visualizations that
emphasize progress at different course granularities, and study whether specific
visualizations are more suitable for specific course sub-populations. More specif-
ically, we explore how adult and adolescent students respond to visualizations
that highlight weekly progress and course-long progress.
This article is structured as follows. First, in Sect. 2 we provide an overview
of students’ orientations and mindsets as well as discuss point and progress
visualizations. Then, in Sects. 3 and 4 we explain the details of the study and
its results, respectively. In Sect. 5 we discuss the results and lastly, in Sect. 6 we
conclude this work and outline possible future research.
2 Related Work
In this section, we visit two relevant streams of research. First, we briefly discuss
self-regulation in adults and adolescents. Then, we discuss previous work on
motivational visualizations.
2.1 Students’ Mindsets as Factors of Success

This stream of research aligns with human self-regulatory processes, which
broadly encompasses the ability of an individual to control and evaluate their
behavior during learning. Caprara et al. [3] studied how changes in self-regulatory
efficacy affects students’ grades. They found that as the efficacy declined, the
students’ grades lowered. This in turn led to a higher chance of dropping out
from studies. Their dataset consisted of 412 students, and the study was con-
ducted as a longitudinal study over a period of ten years – the participants were
followed between the ages of 12 and 22.
In her research, Dweck observed that individuals who believe that intelligence
is fixed, or that success is based on innate ability, are more likely to stop working
during setbacks or when struggling than those who believe that intelligence can
grow through effort [4]. Dweck popularized this belief using the term “mindset”.
Later, different types of interventions that emphasize growth of knowledge over
fixed knowledge have been developed.
For example, O’Rourke et al. [12] developed two versions of a math video
game called “Fractions” to study the effects of growth mindset. Their experi-
mental version of the game promoted students’ learning with commentary that
praised their intelligence, whereas the control group got more neutral feedback
praising their mathematical skills. They found that the experimental group were
more persistent as they played for a longer period of time and completed more
levels than the control group. They also started to display behavior coherent
with a growth mindset by using strategies they learned from the game. Another
result of the study was that lower performing students were more encouraged to
keep playing in the experimental version than in the control version.
The goal is to provide the students visualizations that best support self-
regulation. This in turn would lead the students to the correct mindset which
prevents quitting.
Adolescent and Adult Student Attitudes Towards Progress Visualizations 17
Steinberg et al. [15] found that young adults’ future orientations and delay
discounting continued to develop from childhood all the way into their mid-
twenties. The research compared 10 to 30 year old individuals and indicated,
when compared to their counterparts, that younger individuals are more likely
to choose a smaller reward sooner than a larger reward later. This supports our
study design.
2.2 Point and Progress Visualizations

Overall, point and progress visualizations fall under multiple umbrellas. They
are an integral part of gamification research that seeks to bring game-like ele-
ments into educational contexts as well as more generic educational principles
of transparency where students’ progress should be made visible.
Leppänen et al. [8] studied the effect of progress visualization on student
engagement. They found that, at least initially, a simple progress bar significantly
increased the quantity of completed exercises. Their system displayed the bar at
the bottom of the screen, and the bar became full when the user had answered
three consecutive multiple choice questions correctly. However, the arbitrary
nature of this goal meant that their bar simply created an illusion of progress –
which was sufficient for increasing students’ effort – rather than depicting any-
thing more meaningful.
Loboda et al. [9] developed and studied a form of open learner modelling
which is comparable to our progress visualizations, though more clearly split up
by subject. However, in their study, the use of the system was entirely optional
so any apparent effect might simply be due to better students being more likely
to use the system.
Another study of open learner modelling was conducted by Bull et al. [2].
They used multiple visualization types for the same student, and the students
could choose which visualizations were visible to them. The available visualiza-
tions included skillometers, tables, stars, smileys, gauges, histograms, networks,
radar plots, word clouds and treemaps. These visualizations had information
about students’ knowledge and skills on the subject. They found that students
prefer simpler visualizations such as skillometers and stars over complex visual-
izations such as treemaps and word clouds.
Santos et al. [14] studied similar use of visualizations. Their research had a few
differences from ours. Their study was iterative. In the first phase they collected
feedback from teachers. The teachers valued simplicity and intuitivity, which
is in line with Bull et al.’s results [2]. After modifications between evaluations,
they provided the visualizations to students. They found that their demographic,
which consisted of 18 to 20 year old students, valued the visualizations. Another
difference to our research was that they showed the students their peers’ progress,
which the students found positive.
Visualizations have also been used in education to present students’ progress
to teachers [5,7]. Park and Jo [13] studied different usages of learning analytics
dashboards for students, teachers or both. Based on the information they col-
lected from previous studies of dashboards, they developed their own dashboard
18 O. Aarne et al.
for both students and teachers, with the focus on students’ self-regulation in the
variables shown in the dashboard. They developed their dashboard iteratively
similarly to Santos et al. [14] by interviewing students and improving the dash-
board based on the feedback. Their dashboard also allowed students to see each
other’s progress.
Gamification in education [6,10,11] utilizes visualization of students’
progress. For example Holman et al.’s [6] learning management system Grade-
Craft shows students their course scores as a progress bar. Additionally, they
can get badges as they learn new concepts which are also shown in the progress
bar. GradeCraft shows useful visualizations for teachers as well. They can see
which badges have been completed, and which are in progress. They also see
how the students predicted their success during the course and if their progress
corresponds the prediction.
3 Methodology
In this section, we first explain the context of our study in Sect. 3.1. Then,
we describe the survey the students were given and the students’ answers, i.e.
the data for our study in Sects. 3.2 and 3.3. Lastly, our research questions are
outlined in Sect. 3.4.
3.1 Context
The data for this study comes from a CS1 course series split into two seven-week
courses conducted in the spring of 2016 at the University of Helsinki. The course
series is an introduction to object oriented programming with Java with content
similar to most CS1 courses globally. Since most CS majors take these courses
in the fall, our sample mostly consists of non-CS majors.
The course uses a blended online material with interspersed theory sections
and assignment sections. For the purposes of this study, a visualization compo-
nent was added to the online material.
In the study, the students were randomly divided into two groups, which were
shown separate visualizations of their progress. One group was shown a series of
bars that corresponded to the percentage of completed exercises for each week
of the course, and the other group was shown a line plot that emphasized the
growth of students’ skills throughout the course. The visualization aggregated
students’ scores over the course, showing a steady growth as students’ progressed
throughout the course.
From this point onward, we refer to the visualization with the weekly bars
as the Weekly visualization, and the visualization with an emphasis on course-
long growth and effort as the Growth visualization. Two examples of the Weekly
visualization are shown in Fig. 1, while Fig. 2 shows two examples of the Growth
visualization.
The visualization was implemented as a floating element that was constantly
available in the lower right corner of the course material. As the students worked
on the course, they were able to hide the visualization temporarily. Regardless
of whether the student had previously hidden the visualization, the visualization
was visible to the student when the material page was opened again.
Fig. 1. Examples of the Weekly visualization, where the percentage of completed

assignments for each week of the course are shown as a bar chart. The student on
the left did fairly well, whereas the student on the right could not quite keep up but
did not give up.
Fig. 2. Examples of the Growth visualization where aggregate assignment score is

displayed on the x axis and time is shown on the y-axis. On the left is a perfect score,
and on the right the student slowly fell behind and gave up.
3.2 Survey
At the start of the final exam of the course, the students were asked to fill out a
survey about the course. As an incentive, they were offered an extra point toward
the exam score. All of the survey questions relevant to our study were Likert
items scaled from 1–7, where a low number indicated agreement and vice versa.
A zero option was also included if the students felt the item was not applicable to
them. The items for this study included statements about whether the students
saw the progress visualization, whether they liked the visualization they saw
and whether they felt motivated by it. The distributions of the responses to the
questions are shown in Figs. 3, 4, 5 and 6.
20 O. Aarne et al.
3.3 Data
Our study uses data from two subsequent programming courses that are typically
taken together. In our dataset each student in a course makes up a single data
point. Therefore students who took both courses contributed two points of data.
Our survey from the course exams provided data from 89 and 83 students for
each course respectively. For the study, we only included students who agreed
that they saw the visualization, i.e. whose responses were between 1 and 3 on
the 7-point Likert scale. This narrowed our dataset down to a total of 118 data
points. We had to exclude 27 data points due to missing information on the
participants’ ages, leaving us with a final dataset of 91 data points.
We divided the data into two groups: adolescents and adults. We consider
adults to be students born before the year 1995, i.e. students who turned 22 the
year of the study. This cutoff was chosen since it formed quite even groups and
kept the age of the adolescents reasonably low. A later cutoff year would have
raised the adolescent group’s mean age to over 20. With our chosen cutoff the
adolescents were 18–20 year olds, while the adults’ age ranged from 21 to 34
years. The medians, means and standard deviations of the ages for the different
groups are reported in Table 1.
There were 34 adolescents and 57 adults. Of all the individual students, 17
only took the first course, 20 only took the second course and 27 took both.
For the students who took both courses, the visualization was switched between
courses, which means that 27 students reported their experiences with both
visualizations. The number of students in different age groups for the two visu-
alizations can be seen in Table 1.
Table 1. The group sizes for the different age and visualization groups, along with the
medians, means and standard deviations of the groups’ ages.
Age group Weekly vis. Growth vis. Median age Mean age σ of age
≥21 27 30 23 24.11 3.64
<21 11 23 20 19.56 0.56
Total 38 53 21 22.41 3.64
3.4 Research Questions
We have the following two research questions for this work:
– RQ1. Does the type of visualization affect student scores?

– RQ2. Do participants of different ages respond differently to the different
progress visualizations?
With the first research question, we are interested in studying whether the
two visualizations affect students’ scores differently, that is, whether the type of
visualization matters. With the second research question, we examine whether
a student’s age affects how the student is affected by the visualization.
4 Results
To get a general overview of the results and distributions of the relevant survey
responses, we visualize them in Sect. 4.1. We then analyze the results in greater
detail, and seek to answer the research questions in Sects. 4.2, 4.3 and 4.4.
To answer research question 1, we normalized the total assignment scores
for each student and calculated the means and standard deviations for each age
group by visualization, reported in Table 4.
For research question 2 we calculated means and standard deviations of the
responses to the survey questions “Did you like the visualization?” and “Did the
visualization motivate you to complete more assignments?”, which are reported
in Tables 2 and 3 respectively.
4.1 Questionnaire Responses

The distributions of the questionnaire responses for each age and visualization
group are reported above in Figs. 3, 4, 5 and 6.
Fig. 3. The questionnaire response distribution from the adults who saw the weekly
visualization.
Fig. 4. The questionnaire response distribution from the adults who saw the growth
visualization.
Fig. 5. The questionnaire response distribution from the adolescents who saw the
weekly visualization.
Fig. 6. The questionnaire response distribution from the adolescents who saw the
growth visualization.
22 O. Aarne et al.
4.2 Preference
The different age groups seemed to prefer different visualizations. When asked if
they liked the visualization, the mean response level of adolescents who saw the
Weekly visualization was 1.55, whereas the adolescents’ opinion on the Growth
visualization was only 2.00. This indicates that the adolescents collectively pre-
ferred the Weekly visualization to the Growth visualization. Conversely, the
mean response levels of adults were 2.22 and 1.65 for the Weekly and Growth
visualizations respectively, i.e. the adults liked the Growth visualization more
than the Weekly visualization.
There is very little difference between the standard deviations for the Growth
visualization. For the Weekly visualization the standard deviations are signifi-
cantly different between the two age groups: 0.69 for the adolescents and 1.72
for the adults. These values are reported in Table 2.
Table 2. Likert scale responses to the question “Did you like the visualization?”. The
responses were between 1 (completely agree) and 7 (completely disagree).
Age Weekly vis. Growth vis.

Mean σ Mean σ
≥21 2.22 1.72 1.65 1.28
<21 1.55 0.69 2.00 1.38
4.3 Motivating
The age groups were also more motivated by their preferred visualization. Adults
reported being more motivated by the Growth visualization. When asked to
report if the visualization affected their motivation to complete assignments, the
mean response levels were 3.04 for the Weekly visualization and 2.19 for the
Growth visualization. Similar to the preference question, the effect was reversed
for adolescents. Their mean response levels were 2.27 for the Weekly visualization
and 2.96 for the Growth visualization.
The differences in standard deviations follow the same trend that appears
in the preference subsection, only to a lesser extent. The Growth visualization’s
Table 3. Likert scale responses to the question “Did the visualization motivate you
to complete more exercises?”. The options were between 1 (completely agree) and 7
(completely disagree).

Mean σ Mean σ
≥21 3.04 2.01 2.19 1.64
<21 2.27 1.79 2.96 1.75
standard deviations are not significantly different, whereas the Weekly visual-
ization’s standard deviations are 2.01 for the adults and 1.79 for the adolescents,
i.e. there is a slight difference. Although the standard deviations of the Weekly
visualization group differ for both of the questions, the difference is significantly
higher for the preference question. These values are reported in Table 3.
4.4 Points
The normalized assignment scores for the two different visualizations and age
groups are reported in Table 4. We see a similar effect, albeit weaker, as with
motivation and liking: younger students complete slightly more assignments
when shown the Weekly visualization (86% completed vs 80% completed) and
older students complete marginally more assignments when shown the Growth
visualization (84% completed vs 81% completed).
Table 4. Average scores for the different age and treatment groups normalized to be
between 0 and 1.

Mean σ Mean σ
≥ 21 0.80 0.13 0.84 0.10
< 21 0.86 0.08 0.81 0.09
5 Discussion
Our results indicate that the long-term focused visualization received more pos-
itive responses from adult students and was associated with higher scores com-
pared to the alternative. Similarly, the visualization which emphasizes weekly
progress received more positive responses and resulted in higher scores in the
younger group, compared to the alternative.
Therefore, the answer to our first research question, “Does the type of visual-
ization affect student scores? ”, is affirmative. Moreover, as younger and older stu-
dents preferred different visualizations, a more specific answer that also answers
our second research question, “Do participants of different ages respond differ-
ently to the different progress visualizations? ”, is the following: “Depending on
the age of the student, different progress visualizations affect students’ scores dif-
ferently”. However, the differences between the visualizations were subtle. This
is especially the case when comparing our results to Leppänen et al. [8], who
found a much more intense effect. A number of relevant differences may explain
this. Their study used simple multiple choice questions, and kept the progress
bar always visible – the progress bar also indicated when the student had “com-
pleted” a specific task, even though in reality their completion was arbitrary. In
our study, the progress was slower as the assignments were more complex, and
the visualization was only visible in the web-based course material and not in
programming environment where students worked on most of the course tasks.
24 O. Aarne et al.
It is possible that a greater effect could be achieved if the visualization was

shown after every completed assignment, also within the programming environ-
ment in which the students completed their course work.
We found that older students like the Growth visualization that emphasizes
long-term gains more than the Weekly visualization that highlights short-term
progress. These results are in line with those reported by Steinberg et al. [15]
who found that students’ future orientation continues to develop until their mid-
twenties. This suggests that there might be a benefit to adjusting motivational
tools according to the students’ ages, even among legal adults.
Generally, students both liked and were motivated by the two visualizations.
For all groups, the mean response level was under 4 on a scale from 1 (completely
agree) to 7 (completely disagree). Thus, our results indicate that students like
both visualizations, and the differences in our results are only in the magnitude
of the liking. This is in line with previous work by Bull et al. [2], who found that
students enjoy simple visualizations, as both of the visualizations in this study
can be considered simple. On the other hand, it is possible that students who
did not like the visualization turned the visualization off, and therefore answered
that they did not see the visualization – such students were excluded from this
study.
Another reason for the younger students thinking more favourably of the
Weekly visualization might be that it is more familiar to them. Progress bars
are commonly used for depicting various elements such as skill proficiency and
enemy health in video games, and presumably younger students spend more time
on them on average compared to older students.
In a broader context, our results indicate that in order to increase students’
learning, it might be more effective to personalize visualizations depending on
students’ demographic factors such as their age. Even subtle improvements in
students’ success could be beneficial in e.g. programming courses that suffer from
notoriously poor pass-rates [1].
5.1 Limitations
In this subsection, we acknowledge the limitations of our work. As our study

is based on a small population with rather small differences in the outcomes,
none of the results are statistically significant, despite the visual analysis that
indicates otherwise. Therefore, this study should be treated more as qualitative
first insight into this topic rather than a quantitative examination of the effect of
different visualizations on learning. Comparisons between adult and adolescent
learners are always difficult as the age is rarely the only difference; older students
are often at very different points in their lives. Due to our method of splitting the
students into only two age groups of under 21-year-olds and at least 21-year-olds,
the younger group is much more homogeneous than the older one. For example,
there is likely a larger difference between a 22-year-old student and a 32-year-old
student compared to a 22-year-old student and a 20-year-old student, but in our
study the 22-year-old would be in the same group with the 32-year-old student.
Our study also lacked a control group of students who saw no visualization
at all. As such, we can not say how much the existence of a visualization affected
students performance over a baseline – we can only compare the effect between
the two visualizations. However, because most of the students we studied said
they liked and were motivated by the visualization they saw, we can assume that
the visualizations did not have a negative impact on the students’ scores. That
is, we may assume that a control group would have performed more poorly.
6 Conclusions
In this article we studied how different age groups responded to being shown dif-
ferent visualizations of their course progress – overall, students liked the visual-
izations and they can be a useful tool for educators in keeping students motivated
and engaged. However, our preliminary results suggest not all such visualizations
are made equal. Out of our two progress visualizations, the one that emphasized
long term growth was better liked and promoted better scores among the older
age group, whereas the younger age group performed better with a bar plot
visualization that emphasized each week’s progress individually. The differences
were rather subtle, however, and were not statistically significant.
In the future, we would like to further test our findings on a course with
more students. It would also be valuable to have a control group who were not
shown any visualization to study how these visualizations compare to not having
a visualization at all. In addition, we want to study how increases in the number
of completed exercises translates to actual acquired skills, for example whether
there are differences in exam scores between the different visualization groups.
References
1. Bennedsen, J., Caspersen, M.E.: Failure rates in introductory programming. ACM
SIGCSE Bull. 39(2), 32–36 (2007)
2. Bull, S., Ginon, B., Boscolo, C., Johnson, M.: Introduction of learning visualisations
and metacognitive support in a persuadable open learner model. In: Proceedings
of the Sixth International Conference on Learning Analytics & Knowledge, pp.
30–39. ACM (2016)
3. Caprara, G.V., Fida, R., Vecchione, M., Del Bove, G., Vecchio, G.M., Barbaranelli,
C., Bandura, A.: Longitudinal analysis of the role of perceived self-efficacy for self-
regulated learning in academic continuance and achievement. J. Educ. Psychol.
100(3), 525 (2008)
4. Dweck, C.S.: Mindset: The New Psychology of Success. Random House Incorpo-
rated, New York (2006)
5. Dyckhoff, A.L., Zielke, D., Bültmann, M., Chatti, M.A., Schroeder, U.: Design and
implementation of a learning analytics toolkit for teachers. J. Educ. Technol. Soc.
15(3), 58 (2012)
6. Holman, C., Aguilar, S., Fishman, B.: Gradecraft: what can we learn from a game-
inspired learning management system? In: Proceedings of the Third International
Conference on Learning Analytics and Knowledge, pp. 260–264. ACM (2013)
26 O. Aarne et al.
7. Jacovina, M.E., Snow, E.L., Allen, L.K., Roscoe, R.D., Weston, J.L., Dai, J.,
McNamara, D.S.: How to visualize success: presenting complex data in a writing
strategy tutor. In: EDM, pp. 594–595 (2015)
8. Leppänen, L., Vapaakallio, L., Vihavainen, A.: Illusion of progress is moar addictive
than cat pictures. In: Proceedings of the Third ACM Conference on Learning@Scale
(L@S 2016), pp. 133–136, New York, NY, USA. ACM (2016)
9. Loboda, T.D., Guerra, J., Hosseini, R., Brusilovsky, P.: Mastery grids: an open-
source social educational progress visualization. In: Proceedings of the 2014 Con-
ference on Innovation & #38; Technology in Computer Science Education (ITiCSE
2014), p. 357, New York, NY, USA. ACM (2014)
10. Morrison, B.B., DiSalvo, B.: Khan academy gamifies computer science. In: Pro-
ceedings of the 45th ACM Technical Symposium on Computer Science Education,
pp. 39–44. ACM (2014)
11. O’Donovan, S., Gain, J., Marais, P.: A case study in the gamification of a university-
level games development course. In: Proceedings of the South African Institute for
Computer Scientists and Information Technologists Conference, pp. 242–251. ACM
(2013)
12. O’Rourke, E., Haimovitz, K., Ballweber, C., Dweck, C., Popović, Z.: Brain points:
a growth mindset incentive structure boosts persistence in an educational game. In:
Proceedings of the SIGCHI Conference on Human Factors in Computing Systems,
pp. 3339–3348. ACM (2014)
13. Park, Y., Jo, I.-H.: Development of the learning analytics dashboard to support
students’ learning performance. J. UCS 21(1), 110–133 (2015)
14. Santos, J.L., Govaerts, S., Verbert, K., Duval, E.: Goal-oriented visualizations of
activity tracking: a case study with engineering students. In: Proceedings of the
2nd International Conference on Learning Analytics and Knowledge, pp. 143–152.
ACM (2012)
15. Steinberg, L., Graham, S., O’Brien, L., Woolard, J., Cauffman, E., Banich, M.:
Age differences in future orientation and delay discounting. Child Dev. 80(1), 28–
44 (2009)
3D Printing as Medium for Motivation
and Creativity in Computer Science Lessons
Petra Kastl, Oliver Krisch, and Ralf Romeike(B)
Computing Education Research Group,

Friedrich-Alexander-Universität Erlangen-Nürnberg,
Martensstraße 3, 91058 Erlangen, Germany
{petra.kastl,oliver.krisch,ralf.romeike}@fau.de
Abstract. 3D printers are a modern technology which is new to most

students and can be used to motivate them. However, if the use of a
3D printer is limited to the downloading and printing of prefabricated
models, the students will learn little about computer science. In this arti-
cle, we present an approach for learning basic programming skills, which
is based on the programming of three-dimensional objects using turtle
graphics in Beetle Blocks. In addition, there should be an interdiscipli-
nary link to the functions taught in mathematics teaching. The motiva-
tion was the creation of a 3D figure based on mathematical functions,
its creative elaboration and its 3D printing at the end of the sequence.
The experience shows that students are inspired by the idea, but the link
with mathematical foundations is a challenge.
Keywords: 3D printing · Programming · Junior high · Beetle Blocks ·

Creativity
1 Introduction
For a long time, technological progress with respect to school classes was pri-
marily driven by better graphics performance and was used for motivating stu-
dents in the school, e.g. in the field of computer graphics. In recent years, 3D
printers have been a new technological development, usually not (yet) found in
the typical household. Nevertheless, it can fascinate students, their colleagues
and parents alike with amazing results. Therefore, 3D printers are purchased
at various schools to provide the impressive possibilities of 3D printing to stu-
dents. In practice, however, the use of 3D printers is often limited to configuring
the printer and printing out prefabricated models downloaded from the Inter-
net, which is supported by [2]. As of yet, few convincing examples have been
published that make use of the interdisciplinary potential of 3D printing in com-
puter science lessons. In the following, we present a teaching example in which
students are motivated by creating and printing algorithmically-generated 3D
models as an introduction to programming. The students experiment with con-
cepts they know from their mathematics lessons. Furthermore, the application
https://doi.org/10.1007/978-3-319-71483-7_3
28 P. Kastl et al.
of these mathematical concepts in combination with simple algorithmic struc-

tures makes possible the creation of impressive figures and thus opens up the
application context outside mathematics.
2 Geometric Shapes and Algorithmic 3D Models

2.1 Turtle Graphics as the Basis for Algorithmic 3D Models
In the 1980s, Seymour Papert introduced Logo, a programming environment

that allows children to explore geometric shapes by drawing them on the screen
using simple programming instructions. In his learning theory called construc-
tionism, Papert [5] emphasizes the role of creating concrete artifacts that are
personally meaningful and can be shown, tested and admired by others. The
following examples are based on the corresponding turtle graphics approach,
but extended into three-dimensional space: The output of the 3D printer, or its
virtual representation, corresponds to the traced path of the turtle in Logo or to
the pen of an object in Scratch. Many everyday objects can be constructed from
mathematical objects and shapes and are suitable as a basis for turtle graph-
ics. The starting point in our teaching example was a saltshaker. Its shape is
x2 y2
equivalent to a hyperboloid with the formula 5.5 2 − 2.52 = 1 (see Fig. 1). The
idea was to build similar three-dimensional objects, which are based on mathe-
matic functions, and print them with a 3D printer. To implement the code, the
block-oriented programming language Beetle Blocks [3] was used.
Fig. 1. Saltshaker - hyperboloid as a template for 3D objects.
The saltshaker is based on the underlying shape of a circle, which serves as

the starting shape in our example. The pupils generated the circle line√with the
help of the formula r2 = x2 +y 2 . The formula was transformed to y = ± r2 − x2 ,
which divided the circle by drawing two semicircles. The pupils could now draw
a cylinder by slowly increasing the height z. In a second step, they then learned
3D Printing as Medium for Motivation and Creativity 29
to change the radius of the individual circles in relation to the height. This
was firstly achieved exemplarily by simple mathematical operations, which were
then replaced by more complex operations during the course of the lesson. The
advantage of building the object from individual layers is that the 3D printer
also builds its objects by adding individual layers. Thus, the pupils were able to
see in advance how the whole object emerges from individual layers (cp. Fig. 2).
Fig. 2. Printing of a Beetle Block 3D model with an ultimaker 2.
2.2 Visual Programming with Beetle Blocks

With the introduction of Scratch [4], the entry-level for beginners has been
significantly reduced, which is particularly beneficial for pupils at junior high.
By visualizing the programmed objects (sprites) on the stage, “programming”,
in the sense of Paperts’ constructionism, is immediately experienced, the multi-
medial bandwidth allowing the pupils to implement personally significant ideas
and to directly follow the consequences of their programming decisions. The
representation of the instructions as blocks also avoids syntax errors and the
available programming constructs are immediately available. Just as Strecker [6]
reported on the use of graphical programming languages in high school, due to
its intuitive and easy handling, the visual access is also appealing in junior high.
Koschitz and Rosenbaum [3] now transfer the concepts and user interface of
Scratch, as well as the idea of turtle graphics, into three-dimensional space using
Snap [1]. With Beetle Blocks, three-dimensional objects can be programmed
based on algorithmic structures (see Fig. 3). The results can subsequently be
exported as a 3D model and can be printed out using a 3D printer. Algorith-
mic, mathematical and artistic objects are thus becoming directly tangible. In
the classroom, we have had good experiences with designing jewelry, everyday
objects such as vases and bowls, as well as artistic figures. The potential of a
30 P. Kastl et al.
lesson aiming towards these goals is that even those students who are not pri-
marily interested in programming can be motivated by the inventive creation of
products typically not expected to be encountered in computer science.
Fig. 3. Beetle Blocks.
3 Implementation
3.1 Context and Background
In tenth grade at Bavarian junior high, at the latest, students will learn about
the basic structures of sequence, conditionals, and loops in the “Modeling and
Coding Algorithms” module of the curriculum for the subject of “information
technology”, which includes computer science. Another point of the curricu-
lum is the implementation of algorithms with a suitable programming tool. The
motivation of the students is a major concern in these lessons. Correspondingly,
the programming languages used for learning have changed over the course of
time. For example, the Pascal programming language was previously used at our
school. In order to increase motivation, we moved to Delphi. With Delphi, the
students programmed a pocket calculator or a vending machine at the end of the
teaching sequence, applying the newly acquired skills in a more complex context.
Such a self-programmed pocket calculator motivated some, but did not inspire
enthusiasm in the student group. The reasons for this were twofold: on the one
hand the very static interaction with the environment, and on the other hand the
high error rate due to syntactic errors. This has changed with the introduction
of the visual programming language Scratch. With this programming language,
the students were able to achieve an appealing result very quickly and without
syntactic errors. The first small games were programmed very quickly and could
be taken home. Working with Beetle Blocks was similar to Scratch. With a few
commands, the students could quickly move the beetle over the screen. This
caused amusement and curiosity in the group.
Fig. 4. Snail of equilateral triangles
3.2 Teaching Objectives and Lesson Structure
The aim was for students to use simple basic mathematical forms, e.g. circle,
square or triangle, to develop three-dimensional objects and to program them.
As a template, the students were able to use familiar subjects from their daily
lives, while being able to freely choose their 3D objects, since they had different
ideas ofwhich objects to create. One of the goals was to give the students the
opportunity to use their own creativity independently and implement their own
ideas. For this purpose, the students had to draw on their knowledge of geome-
try. In addition to the well-known theorem of Pythagoras, the basic forms also
included other basic knowledge, for example, the division ratio of the heights in
an equilateral triangle. With this knowledge, the basic shapes were created very
quickly (cp. Fig. 5). For the third dimension, mathematical functions were used:
With the help of sine, cosine and other functions, the size of the basic shapes was
changed as a function of the height to give the objects a nicely curved shape. In
Fig. 4 we see a colored snail, which is composed of equilateral triangles. In the
snail, the side length was gradually reduced in size.
The lesson is divided into three phases:
1. Learning the algorithmic basic structures of sequence and loops and applica-
tion/exercise with the programming language Beetle Blocks. In this case, the
pupils produce simple geometric figures, such as e.g. rectangles or squares
(3 double lessons)
2. Expansion into the third dimension, by the students building the simple ge-
ometric figures as a tower (1 lesson)
3. Mathematical modification of the tower (3 double lessons, cp. Fig. 7))
32 P. Kastl et al.
Fig. 5. Programming of basic geometric shapes.
3.3 Course of Instruction and Observations

The example in focus was taught in a 10th grade class with pupils from the
economic branch of a junior high school. Therefore, the initial enthusiasm of
the students to learn programming was lacking. Linking the content with the
“beloved” subject of mathematics did not contribute to the motivation. This was
shown, for example, in the statement of student Clara: “I don’t understand math,
and now I have to learn how to program it.” Correspondingly, the start of the
lesson was difficult. The students were to learn to use the term “sequence”. For
this purpose, we programmed simple geometric figures, for example, a rectangle.
The programming of an equilateral triangle was somewhat more elaborate, since
the coordinates of the third corner had to be calculated with the help of the
Pythagorean theorem. The fourth lesson was somewhat easier for the pupils
because only the variable z, which had not yet been used, was added. Up until
this point, they still had not used repetition. During the construction of the
tower, the pupils programmed the height of their tower by repeatedly increasing
the variable z with a counting loop.
Starting from the fifth lesson, the motivation of the pupils increased signifi-
cantly. The teacher brought a 3D-printed object and promised the students their
objects would also be printed with the 3D printer. After the constructive intro-
duction, in the sense of a bottom-up approach, further teaching now pursued
an experimental-deconstructing approach. For this purpose, different methods
were provided to the pupils in the form of given blocks, which, in addition to the
Fig. 6. Student results: Variations of the towers.

Fig. 7. Construction and variation of a triangle-based vase.

34 P. Kastl et al.
Fig. 8. Vase, art object and snail from the 3D printer.
blocks developed by the pupils, provided the possibility for the geometric figures
to still be rotated by an angle α.
The reason for providing the blocks was that the students had not yet learned
the underlying mathematical concepts in their mathematics lessons. In addition,
the time given was seven-hours, which can be considered short, and the students
should be given the opportunity to spend more time developing their own 3D
objects. In doing so, they elaborated the previously learned method of building
the individual objects from layers. In the course of time, the attitude of Clara
also changed. With statements like “Is there another mathematical function that
I can try?” Clara wanted to try ever more variations of her tower. Thus, not only
were simply turned towers produced, but also, for example, small bowls. Other
students built nested objects (see Fig. 6).
After all, one of the most beautiful models was created by Clara. A snail-shell,
which consisted of turned, equilateral triangles with a decreasing side-length
towards the top (see Fig. 8). All results were then printed with a 3D-printer and
could be taken home by the students.
Summarizing, the students achieved the following milestones:
1. Creation of two-dimensional geometric shapes using sequences and variables
as the basis.
2. Creation of uniform three-dimensional shapes as an aggregate of two-
dimensional shapes (slices) using loops and procedures.
3. Experimenting with varying parameters by applying trigonometric functions
to the side length of the basic geometric shapes, using a given block.
4. Experimenting with rotation of the basic geometric shapes.
5. Inclusion of additional parts, such as a foundation.
4 Conclusion and Outlook

As a conclusion, the students’ motivation increased when it became clear that the
3D objects they were programming were to be printed three-dimensionally. Their
involvement in learning programming was comparable to that of the students
learning to program with Scratch. Also, regarding the speed of learning, there
was no difference to the Scratch group. With the first basic geometric figures,
things became more difficult. The students were able to calculate the coordinates
of the corners of a square without any problems, but with an equilateral triangle,
some problems arose. Here, a different approach should be found and tested
with simpler mathematical methods. The use of mathematics was significantly
different from the way students are used to solving problems in their mathematics
lessons. Perhaps, students from the mathematical branch, or grammar school,
have fewer difficulties. A group of two students did something different: Instead
of calculating the coordinates, they have constructed an equilateral triangle on
a piece of paper, placed it in the origin of the coordinate system, and read the
coordinates of the equilateral triangle from the scales. However, in the third
dimension, the two students could not change the size of the sides. For the
others, the transition to the third dimension was easier: Since the basic figures
were programmed in such a way that their size depended on the length of the
side, this could be easily changed using mathematical functions. The pupils’
creativity was given free rein. They were able to apply the functions they had
learned so far. In this step, a few students varied the thickness of the lines drawn
by the beetle. Another group of students also changed the color of the drawn
lines with the height. Two students provided their 3D object with a floor plate
to get a vase. This part gave the pupils lots of fun. A link to other modules
from the curriculum, for example with Arduinos (module: micro-controller) as a
control for an artistic lamp (see Fig. 8), would be a suitable next step.
For the repetition of the teaching sequence, it is recommended to try creating
the basic figures with simpler mathematical methods. The linking of height with
mathematical functions worked well. Here, the students have the opportunity
to see that mathematics is not only an abstract, theoretical subject, but that
it also has an application to give space to one’s own creativity and to develop
new objects. Another new starting point would be to subdivide real objects (see
saltshakers) into mathematical basic objects and to reprogram them.
References
1. Harvey, B., Mönig, J.: Bringing “no ceiling” to scratch: can one language serve kids
and computer scientists. In: Proceedings of Constructionism, Paris (2010)
2. Shewbridge, R., Hurst, A., Kane, S.K.: Everyday making: identifying future uses
for 3D printing in the home. In: Proceedings of the 2014 Conference on Designing
Interactive Systems, pp. 815–824. ACM (2014)
3. Koschitz, D., Rosenbaum, E.: Exploring algorithmic geometry with “Beetle Blocks”:
a graphical programming language for generating 3D forms. In: 15th International
Conference on Geometry and Graphics Proceedings, Montreal (2012)
36 P. Kastl et al.
4. Maloney, J., Resnick, M., Rusk, N., Silverman, B., Eastmond, E.: The scratch pro-
gramming language and environment. ACM Trans. Comput. Educ. (TOCE) 10(4),
16 (2010)
5. Papert, S.: Mindstorms-Kinder, Computer und Neues Lernen. Birkhäuser Verlag,
Basel (1982)
6. Strecker, K.: Graphical programming languages in high school (Grafische Program-
miersprachen im Abitur). In: Gallenbacher, J.(Hrsg.): INFOS 2015: Informatik all-
gemeinbildend begreifen (16. GI-Fachtagung Informatik und Schule, Darmstadt).
Bonn: Köllen (2015)
Teaching Materials in Informatics for Lower
Secondary School Blind Students
Ľudmila Jašková(&) and Mária Stankovičová
Comenius University, Bratislava, Slovakia

jaskova@fmph.uniba.sk, stankovicova@cezap.sk
Abstract. This paper deals with teaching materials in informatics designed for
lower secondary school blind students. First, we describe our motivation to
develop special teaching materials for blind students. The second chapter pro-
vides an overview of the available computing textbooks for primary and lower
secondary education. In the third chapter we explain the way of using computers
by the blind to justify the need for teaching materials specially designed for this
category of students. Further, we offer a detailed description of the form and
content of teaching materials developed by the CEB project (Computing Edu-
cation for Blind). In conclusion, we describe our research aimed at the verifi-
cation of suitability of the created materials.
Keywords: Informatics education Blind Teaching material Programming

Accessibility
1 Introduction
In recent years, many countries have been directing their efforts towards introducing
computer science to pupils at all levels of education in the appropriate volume and
form. In exposing students to computing at school it is important to provide them with
some basic information about the discipline, as is common with other scientific dis-
ciplines, so that later on they will be able to choose whether to pursue this discipline
further in their future education [1]. The CSTA K-12 standards [2] describe computer
science as the study of computers and algorithmic processes, including their
principles, their hardware and software designs, their applications, and their
impact on society.
In Slovakia, informatics became a standard compulsory subject for almost all
grades of primary and secondary schools in 2008 [3, 4]. The content of this subject
includes: basic ICT literacy, information and data handling, communication via ICT,
algorithms, problem solving, user applications, working with the Internet, text, tables
and images, principles of hardware, social aspects of using ICT. Those themes are
adapted to the age of the pupils, with separate curriculums aimed at primary schools
and secondary schools. Since 2011, we have had an opportunity to teach informatics at
a special school for visually impaired pupils, and in recent years our research has been
aimed at the development of suitable teaching materials for teachers of blind pupils.
Research is under way in the framework of the CEB project (Computing Education for
Blind). This project is scheduled for 2016-2018 and funded by our national agency.
https://doi.org/10.1007/978-3-319-71483-7_4
38 Ľ. Jašková and M. Stankovičová
2 Textbooks of Computing
Since the beginning of this millennium, we have been observing efforts in the education
systems of many countries that the content of informatics as a subject has increasingly
evolved from the narrow concept of using a computer as a tool to a more complex
concept of understanding its principles [5]. The subject has been gradually updated,
enabling teachers to focus on the fundamentals and on sustainable knowledge [6]. This
trend is also evident in the changing content of the books for children.
The authors initially focused on the proper use of computers [7–9] (Fig. 1).
Fig. 1. Books about proper use of computers
Later on, there were more publications devoted to programming and computer
science [10–12] (Fig. 2).
Fig. 2. Books about programming
A collection of textbooks titled Creative Informatics was created in Slovakia

between 2005 and 2010. Lower secondary schools (age 11 to 15) were provided with
exercise books on work with graphic and text editors, spreadsheets and the Internet
[13], but also with a textbook on the principles of computer science [14] and workbook
about programming in Comenius Logo environment [15]. Between 2010 and 2013
textbooks were created also for primary schools (age 6 to 10), e.g. [16]. While in our
view, many of these textbooks present a good tool for the teachers of mainstream
students, they are not suitable for the teachers of blind students.
Teaching Materials in Informatics for Lower Secondary School Blind Students 39
3 Textbooks for Blind Students
Blind people use computers in a totally different way than sighted users. They do not
use a mouse, and all the input is entered via a keyboard. Since they use a screen reader,
which reads the content of the screen in a linear way, they cannot get an instant view of
the screen content, as sighted users do.
A different way of using a computer requires a different way of teaching computer
skills and computer science. Unfortunately, our country suffers from a shortage of
qualified informatics teachers of the blind students. While regular IT teachers do not
have the experience of teaching blind learners, teachers from special schools do not
have the experience of teaching informatics. For both, high-quality teaching material is
highly desirable. Since the IT knowledge and computer skills are essential for blind
learners, so that they can continue at higher levels of study and succeed in the labor
market, informatics is the key subject for them.
Since 1993, Sarah Morley Wilkins has written a series of books on Microsoft Win-
dows, helping blind and visually impaired computer users to make effective use of their
PC in the Windows environment (http://www.sarahmorleywilkins.com/). These books,
together with accessible images (Fig. 3), inspired us to create our educational materials.
Fig. 3. A tactile image showing the Windows 7 Start Menu (http://www.sarahmorleywilkins.

com/books/win7/)
Another source of inspiration was tutorials for the blind from companies that
develop software for the blind (e.g. Freedom Scientific) or from organizations defending
the rights of the blind (e.g. American Foundation for the Blind). All of these materials,
however, were designed for adult users, who are highly motivated to learn how to work
with a computer. They do not need a collection of interesting tasks, but a list of keyboard
shortcuts and step-by-step instructions. We have not found textbooks for so young
students (age 11 to 15) with suitable and at the same time engaging tasks to build and
develop their computer skills. Therefore, our goal was to create teaching materials that
offer a set of interesting tasks with graded difficulty to teach pupils in an enjoyable way.
Within the CEB project we focus on the following areas of informatics:
– Text processing – the ability to create a well-structured and clearly formatted text
document should be considered part of basic digital literacy. Blind people should be
able to imagine what is happening with the text when they use various formatting
tools. They should understand the impact of formatting on the understanding of text
content by the reader.
– Sound processing – blind people live mostly in the world of sounds. They have
aural associations with many objects, feelings and situations. When reading text on
the screen, they hear synthetic speech, which is not as natural as recorded sounds.
Understanding synthetic speech is sometimes demanding and tiring. Activities that
allow blind pupils to listen to natural sounds are a pleasant diversion for them.
These activities stimulate their creativity and are motivating and attractive.
– Programming and problem solving – the ability to perform an activity following a
sequence of instructions as well as the knowledge of how to divide a problem into
partial problems are skills that are indispensable for study and a successful life in
21st century.
– Working with mathematical formulas – the ability to understand and use
mathematical formulas is the key to understanding mathematical concepts and
fundamentals of computer science. Most of the mathematical tasks in the lower
grades of secondary education cannot be solved without the ability to use mathe-
matical formulas. Braille mathematical notation is different from regular mathe-
matical notation. If a math teacher does not manage Braille mathematical notation, a
barrier may arise when he or she teaches blind pupils. This barrier may be overcome
by means of Lambda software, which may be especially helpful if blind students
intend to proceed with their studies.
– ICT for communication and information retrieval – the ability to search for
information and its subsequent use is important for learning various subjects as well
as for life in modern society. The same can be said about the ability to communicate
effectively with classmates and teachers using digital technologies.
The creation of teaching materials is based on our experience gained during
teaching informatics to blind and partially sighted pupils [17]. It is our ambition to
allow students to learn in a constructionist way in the spirit of Papert [18]. Con-
structionism means giving students appropriate tasks, topics, problems that would
facilitate the learning process. We provide graded tasks, which each student can solve
at his or her own pace. Apart from the tasks themselves, the data files that students
work with (modify, supplement, remove bugs and the like) present an integral addition
to the material. We believe that creating tasks with respect to the cognitive abilities of
blind students is of utmost importance [19] and that tactile perception may significantly
aid in knowledge acquisition and generation of ideas. Hence, tactile images are also
included in the materials (e.g. tactile pictures illustrating the appearance of different
font formats, tables, windows and the like).
All created teaching materials will be freely available on the web portal, so IT
teachers from both special and regular schools will be able to download and use them.
Teaching materials for thematic areas Text processing and Sound processing have
been completed and are ready for use. We will gradually complete the other materials
so that all of them will be finished by the end of 2018.
3.1 Text Processing

The aim of the teaching material is to familiarize students with basic typographic
principles so that they will be able to use them when editing and formatting text.
The teaching material contains almost 60 tasks complemented with a few dozens of
data files that pupils should use to solve these tasks. Some tasks are intended for the
initial stage of acquiring skills and gaining experience. These tasks also include
explanations of basic concepts, keyboard shortcuts and step-by-step procedures. They
require MS Word 2007 or higher. In addition, the material also includes tasks designed
to practise and automate skills. Such tasks are mainly included in the last chapter,
which contains ideas for larger projects.
A series of 24 tactile images (Fig. 4) show various font, paragraph, and text for-
matting. These images were created for a printer that combines embossed graphics with
inkjet printing.
Fig. 4. A printed version of a relief image illustrating text alignment [20]
Now, let us take a sample task to explain the difference between educational texts
designed for blind students and for sighted students. The aim of the task is to set the
correct heading styles in the document. The screen reader offers keyboard shortcuts to
move between headings in a document, but the heading styles must be set properly. If
so, blind users can get an overview of the text more easily.
Table 1 lists the task assignment for the blind and the instructions for solving it.
Note that the text contains detailed explanation of the task entry, while the instructions
for solving it contain information on how to perform the required operations using the
keyboard.
If the material were intended for sighted pupils, it would contain less text and
would be completed with a picture of the finished text (Fig. 5).
Table 1. Assignment of the task for the blind students
Open document Headings2.docx. The document mentions various African ani-

mals. The headlines do not have the correct headline style so we cannot move be-
tween them. The first line contains the text African animals. Set this line to the
Heading 1 style. The names of the animals that appear on separate lines should be set
to the Heading 2 style. Find them and set them to the correct headline style.
When you set the correct styles to all the headings, go to the top of the document
and move over the headings.
Hints
• Set the headline style by moving to the line you want to set and press Alt + Shift
+ Left Arrow or Alt + Shift + Right Arrow.
• Set the normal style by moving to a line that you want to set and press Ctrl +
Shift + N.
• We can find out the style of the current text by pressing Insert+F.
Fig. 5. Picture of the finished text
Instructions for solving the task would contain information on how to work with the
text using the mouse and would be supplemented by a picture of Style ribbon (Fig. 6).
Fig. 6. Style ribbon in MS Word
3.2 Sound Processing

The aim of the teaching material is to help pupils acquire basic skills for their creative
work with sounds. Specifically, this means acquiring skills to play audio files, to record
and play sounds, to combine two or more sounds, to modify sounds and to learn the
basic terminology needed to process audio information using a computer. The material
contains almost 30 tasks. Instructions for solving these tasks are tied to the use of the
AudaCity editor (http://www.audacityteam.org/). This editor is fully manageable via
keyboard (http://manual.audacityteam.org/man/keyboard_shortcut_reference.html) and
works well with the screen reader.
Fig. 7. Sound of barking dog represented as waves in AudaCity program (tactile image)
The material also includes tactile images illustrating particular environment ele-
ments (such as a volume slider), as well as tactile pictures illustrating the sound as
waves (Fig. 7).
3.3 Programming and Problem Solving

Since we believe that algorithmic thinking is of great importance for the blind, we find
that the programming profession is very suitable for them. Currently, a significant
percentage of blind programmers are working in software companies. We realize that
the sooner the pupils’ interest in programming is appropriately captured, the more
easily they will be able to decide about their further study needed to obtain the nec-
essary qualification. The aim of the teaching material is to acquire basic programming
skills and to understand concepts such as command, sequence of commands, cycle,
conditional command, procedure, variable.
The teaching material will also include unplugged tasks that do not require a
computer. Some of such tasks will require using robotic toys such as the Bee-Bot,
Constructa Bot or a tactile table [21]. We also plan to include algorithmic tasks from
the Beaver contest or their modifications [22, 23].
However, learning programming requires an appropriate programming environ-
ment. The existing languages intended for children are highly visual in nature, both in
manipulating the code (e.g. drag and drop) and in the effect that the code has (e.g.
animation, moving robot). As such, they are not appropriate to students with visual
impairment. With this in mind, our tasks will be tied to the Alan audio programming
environment developed by one of our students of informatics [24]. This environment
allows users to program audio stories (Table 2), but also helps to develop creativity and
to assimilate all the required programming concepts. The Alan environment will be
freely available along with the teaching material.
Table 2. Program in Alan environment
Say(What animal makes this sound?)

Repeat 3 times
Play(donkey)
End of Repeat
Question(Is that a donkey?) Answer: Yes
Say(Excelent!)
Else
Say(No, it is a donkey)
End of Question
Microsoft Research Cambridge is currently developing a physical programming

language Torino (https://hxd.research.microsoft.com/work/torino.php), which is friendly
to blind pupils (Fig. 8). The beta version should be available in autumn 2017. We plan to
buy this language and create teaching material for this product as well.
According to the current timetable, this teaching material should be completed by
the end of 2017. In line with our ambitions to support multiple platforms, this will be
the most extensive teaching material.
Fig. 8. A physical programming language Torino (https://hxd.research.microsoft.com/work/

torino.php)
3.4 Working with Mathematical Formulas

This teaching material will provide practical instructions on the effective use of the
Lambda editor (Linear Access to Mathematics for Braille Device and Audio-synthesis).
This editor was developed as an outcome of the Lambda project (http://lambdaproject.
org/) and was translated into Slovak language at our university in 2006–2007. As a
follow-up, several courses for teachers and blind pupils were held at our Support
Centre. The Lambda editor is a very useful tool because it helps pupils develop their
understanding of mathematics and informatics [25]. Unfortunately, however, there is a
lack of compact teaching material.
The first part of the teaching material under preparation will be aimed at under-
standing the difference between linear text view and graphical math view. In either of
them, the editor responds to keyboard inputs in a different way.
Fig. 9. An example of a series of compensatory functions (linear text view, graphical math
view, compact view)
The following sections will describe a series of compensatory functions (Fig. 9),
whose aim will be to help students learn the basics of how to use these functions
effectively (line duplication, compact structure of expression, calculator, calculation
result insertion, multiple memory buffer, graphical visualization).
One of the major skills is the skill of keeping the right order of keyboard shortcuts.
This skill requires students to understand the meaning of the notation. Keeping the right
order of keyboard shortcuts is necessary for correct graphic visualization and for the
right interpretation by the screen reader.
The last part will deal with exporting files to various formats, especially XHTML.
Based on our previous experience, the best time to start with the Lambda editor is
in the lower secondary school. Later, when pupils acquire basic skills, they will be able
to use the editor at math lessons as an appropriate exercise tool [26].
3.5 ICT for Communication and Information Retrieval

This teaching material will be finished at the end of 2018. The aim will be to teach
pupils how to use tools for communication over the Internet safely and ethically. In
addition to this, they should learn how to look up and use the information needed for
their personal life, study and work in accordance with the copyright law.
A large number of websites are hard to use for blind learners because they do not
meet accessibility standards (https://www.w3.org/WAI/intro/wcag). We plan to use a
training site that satisfies the conditions of accessibility, which will be developed for
this particular purpose. Search-based tasks will focus on verified locations such as
Google search engine, YouTube sound sources, and Wikipedia sources of information.
4 Verification
As we have already mentioned, we are building on our long-term research observa-

tions, which we made at the elementary school for visually impaired pupils. Most of the
tasks used in our materials have already gone through several iterations consisting of
the design, implementation, verification and modification phases. Similar iterations will
be done with each teaching material as a whole. We will verify them in cooperation
with the teachers of blind pupils in the form of qualitative evaluation study [27]. Two
independent reviewers will produce testimonials for each material, and we will
incorporate their comments. In addition, we will analyze a detailed description of
computing lessons, interviews with students, interviews with teachers. Next, we will
analyze the tasks solved by students (predominantly given as projects or tests).
The material on text processing was verified at the beginning of this year during 10
informatics lessons. It can be said that both students and teachers have found the
material easy to use. On the other hand, we have received a number of interesting
responses from students. Based on this feedback, we will have to make minor
adjustments to the text, to tactile images and to data files.
– Some texts in data files are too large. In order to practice the required operation,
shorter texts would be enough.
– It is necessary to modify the content of some data files as they contain terms that
blind pupils do not understand or cannot imagine (such as the names of flowers they
have not encountered yet).
– Some task instructions refer to keyboard shortcuts containing numbers on the
alphanumeric keypad. If pupils press a number on the numeric keypad, this does
not have the desired effect. It is therefore necessary to specify exactly which keys
should be pressed.
We have found that tactile images must form an integral part of the teaching
material. Blind students could not imagine some objects without them, such as a
bulleted list or more intricate nested lists.
Thanks to tactile images, we have noticed some misconceptions. For example,
some students interpreted the double line setting as doubling of the letter size. Simi-
larly, the centered alignment (Fig. 10) was interpreted as lines starting in the middle of
Fig. 10. A blind student encounters a tactile image illustrating text alignment [20]
the page. After using the tactile image, they realized that the distance between the
beginning of the line and the left edge of the page is the same as the distance between
the end of the line and the right edge of the page.
It was a valuable experience for us as well, because we entered the perception of
blind pupils.
5 Conclusions
As all materials will be available on the web portal, we also expect feedback from IT
teachers from different schools. Once all the materials have been created, we plan to
organize a workshop for teachers of blind pupils and experts in the field of education of
the blind both from our country and from the neighboring countries. This seminar will
be yet another platform for exchanging experience and getting feedback.
Acknowledgements. This paper was created as part of a project Computing Education for Blind
(Kega 014UK-4/2016) funded by KEGA grant. Special thanks to all participating students and
teachers for their help and input into the research.
References
1. Armoni, M., et al.: Early education – what does computing have to do with it, and in what
ways? In: Proceedings of the 9th Workshop in Primary and Secondary Computing
Education, London, UK. ACM DL, New York (2015). ISBN 978-1-4503-3753-3
2. Seehorn, D., et al.: CSTA K–12 Computer Science Standards. Computer Science Teachers
Association, New York (2011)
3. Kabátová, M., et al.: Robotic activities for visually impaired secondary school children. In:
3rd International Workshop, “Teaching Robotics, Teaching with Robotics”, Integrating
Robotics in School Curriculum, Riva del Garda (TN), Italy (2012)
4. The national curriculum in Slovakia (in Slovak). http://www.statpedu.sk/clanky/statny-
vzdelavaci-program
5. Jones, S.P., et al.: Computing at school in the UK (2013)
6. Hromkovič, J., Steffen, B.: Why teaching informatics in schools is as important as teaching
mathematics and natural sciences. In: Kalaš, I., Mittermeir, R.T. (eds.) ISSEP 2011. LNCS,
vol. 7013, pp. 21–30. Springer, Heidelberg (2011). https://doi.org/10.1007/978-3-642-
24722-4_3
7. Schwartau, W., et al.: Internet & Computer Ethics for Kids: (and Parents & Teachers Who
Haven’t Got a Clue.). Interpact Press, Seminole (2001)
8. Gimmotty, S.: Computers Don’t Byte: A Beginner’s Guide to Understanding Computers,
2nd edn. (2004)
9. Selfridge, B., et al.: A Kid’s Guide to Creating Web Pages for Home and School. Zephyr
Press, Chicago (2004)
10. Sethi, M.: Game Programming for Teens. Course Technology, Boston (2008)
11. Briggs, J.R.: Python for kids, 3rd edn. No Starch Press Inc, San Francisco (2013). ISBN
978-1-59327-407-8
12. Vorderman, C.: Computer Coding for Kids. Dorling Kindersley Ltd., UK (2014). ISBN
9780241010433
13. Varga, M., Hrušecká, A.: Creative informatics. First exercise book on the Internet 1. SPN,
Bratislava (2006). (in Slovak), ISBN 80-10-00648-3
14. Kalaš, I., Winczer, M.: Creative informatics. Informatics around us. SPN, Bratislava (2007).
(in Slovak). ISBN 978-80-10-00887-2
15. Blaho, A., Kalaš, I.: Creative informatics. First exercise book in programming. SPN,
Bratislava (2005). (in Slovak), 80-10-00019-1
16. Blaho, A., et al.: Informatics for the second grade of primary school. Aitec, Bratislava
(2010). (in Slovak)
17. Jašková, Ľ.: How to start with learning informatics at elementary school for blind pupils. In:
Trajteľ, Ľ. (ed.) Didinfo 2013, UMB, Banská Bystrica (2013). (in Slovak), ISBN
978-80-557-0527-9
18. Papert, S.: Constructionism vs. Instructionism: speech to an audience of educators in Japan
(1980). http://www.papert.org/articles/const_inst/const_inst1.html
19. Jašková, Ľ., Kaliaková, M.: Cognitive abilities of blind pupils of lower secondary education
and their use in teaching computer science. In: Trajteľ, Ľ. (ed.) Didinfo 2016, UMB, Banská
Bystrica (2016). (in Slovak), ISBN 978-80-557-1082-2
20. Vrábľová, M.: Tactile graphics to support teaching informatics to blind elementary school
pupils. Diploma Thesis, FMFI UK, Bratislava (2017). (in Slovak)
21. Jašková, Ľ., Kaliaková, M.: Programming microworlds for visually impaired pupils. In:
Futschek, G., Kynigos, C.: Constructionism and Creativity 2014, Proceedings of the 3rd
International Constructionism Conference, Östereichische Computer, Gesellschaft, Vienna
(2014). ISBN 978-3-85403-301-1
22. Jašková, Ľ., Kováčová, N.: Bebras contest for blind pupils. In: Proceedings of the 9th
Workshop in Primary and Secondary Computing Education, London, UK. ACM DL New
York (2015). ISBN 978-1-4503-3753-3, http://dl.acm.org/citation.cfm?id=2818324&dl=
ACM&coll=DL&CFID=6147908
23. Jašková, Ľ., Kováčová, N.: Contest for blind pupils – universal design of tasks. In:
Proceedings of the Conference Universal Learning Design, Linz 2016, pp. 79–97. Masaryk
University, Brno (2016). ISBN 978-80-210-8295-3
24. Kováč, M.: Programming environment for visually impaired pupils. Bachelor Thesis,
FMFI UK, Bratislava (2016). (in Slovak)
25. Horňanský, M.: Making math accessible for blind in Slovakia through Lambda editor.
Diploma Theses, FMFI UK, Bratislava (2009). (in Slovak)
26. Kaliaková, M., Stankovičová, M., Vitálišová Z.: Informatics competences of visually
impaired pupils and their use in teaching mathematics and physics with Lambda editor. In:
Trajteĺ, Ľ. (ed.) Didinfo 2015, UMB, Banská Bystrica (2015).(in Slovak), ISBN
978-80-557-0852-2
27. Hendl, J.: Qualitative Research. Basic Methods and Applications. (translated into Slovak)
Portál s.r.o, Praha (2005). ISBN 80-7367-040-2
Teachers’ Expectations and Experience
in Physical Computing
Mareen Przybylla1(B) , Finn Henning1 , Carla Schreiber1 , and Ralf Romeike2

1
University of Potsdam, Potsdam, Germany
przybyll@uni-potsdam.de
2
Friedrich-Alexander-Universität Erlangen-Nürnberg, Erlangen, Germany
ralf.romeike@fau.de
Abstract. In recent years, physical computing has grown in popular-

ity for school education. Children and teenagers learn basic concepts of
embedded systems in a creative and motivating manner. We investigate
teachers’ expectations towards and first experiences with physical com-
puting to gain a better understanding of their needs in professional devel-
opment. We developed, conducted and evaluated a qualitative interview
study, identified typical benefits (e.g. motivation, direct feedback and
tangibility), topics (e.g. algorithms and data structures) and challenges
(e.g. acquisition of suitable hardware and necessary preparations) that
teachers associate with physical computing and derive recommendations
for professional development and supporting measures.
Keywords: Physical computing · Interview study · Expectations and

attitudes · Classroom experience · Computer science education
1 Introduction
“Ubiquitous computing enhances computer use by making many computers
available throughout the physical environment, but making them effectively
invisible to the user.” [23] Ubiquitious Computing, as it was described by Weiser
in the early 1990s, has become reality. Embedded and cyber-physical systems
are pervasive in our society, computers are everywhere and often we don’t even
notice their presence. They communicate with their environment and people by
using sensors and actuators and are connected to each other and services over the
Internet. They enhance our lives in many ways: as little helpers in the household,
as control systems in transportation or medicine or as robots to explore areas
that are inaccessible for humans. Both, in industry and science, the development
and advancement of embedded systems embrace a large and important area of
research. The creation of such systems is supported by the large availability of
suitable hard- and software tools for all purposes and experience levels, also for
non-professional use. Many tools, such as Arduino, Raspberry Pi or the BBC
Micro: Bit are aimed at inexperienced and creative developers who use the main
concepts of embedded systems and computer science (CS) in different contexts,
https://doi.org/10.1007/978-3-319-71483-7_5
50 M. Przybylla et al.
for example, for the production of interactive clothing, installations or works of

art. The creative design and development of interactive, physical objects and
systems is also referred to as physical computing. This development is partic-
ularly attractive for the educational sector, as it allows for an introduction to
embedded systems with low barriers. Also, as creative art and design processes
are connected with CS, the common image of computer scientists as ‘nerdy’ and
‘geeky’ can be broken through. Physical computing brings together hardware
and software components and thus connects the virtual world of computers with
the physical world of human experience. In the last few years, various empirical
reports and promising research results on physical computing have been pub-
lished. For example, motivational aspects and the appeal of the subject area
for all learners, regardless of age or gender, are often fostered and investigated
(e.g. [8,12,17]).
With the integration of physical computing into CS curricula, e.g. Eng-
land’s national curriculum [4] or the new CS curriculum of Berlin/Brandenburg,
Germany [20], the demand for professional development for CS teachers has
grown significantly. Thus, we developed design principles for professional devel-
opment and conducted workshops accordingly [15]. From our experience and
discussion with workshop participants, we had the impression that they were
interested in physical computing and saw positive aspects for their classrooms.
However, in follow-up activities, we found that the teachers were very hesitant to
apply what they had learned. Therefore, in this study, we pursue the reasons for
their reluctance and are particularly interested in the challenges that keep them
from implementing projects in their classroom. We describe the development,
implementation and evaluation of an interview study that was conducted in a
two-day workshop with participants with and without physical computing teach-
ing experiences. We present results about benefits and advantages that teachers
attribute to physical computing and topics that they teach using physical com-
puting tools. In particular, we identify challenges and concerns that deter them
from comprehensive implementations in the classroom, discuss possible solutions
and deduce recommendations.
2 Embedded Systems and Physical Computing in School

Physical computing derives its basic concepts from the field of embedded and
cyber-physical systems. Embedded systems (ES) combine hard- and software
components, are embedded into a technical context and fulfill predefined tasks.
They usually control, regulate or monitor a system continuously. For this pur-
pose, in addition to possible user inputs and outputs, sensor data are recorded
and commands are sent to actuators. Embedded systems often have to meet real-
time requirements (cf. [1,22]). Cyber-physical systems (CPS) are networked ES
that communicate among themselves and over the Internet [9, p. 5]. ES and CPS
are the technological basis for many application areas, such as robotics, wear-
able computing or electronic textiles and disciplines, such as physical computing,
interaction design or art that make use of these core technologies in various,
often non-technical contexts (cf. [16]).
Teachers’ Expectations and Experience in Physical Computing 51
Programmable microcontrollers as tools in CS education are an attractive

and promising approach to physical computing. This discipline has evolved
along with an ever-growing community of artists, designers and hobbyists, who
use microcontrollers to create interactive objects and prototypes for these sys-
tems. Sensors, such as brightness or temperature sensors and actuators such as
servo motors or speakers, are used to continually interact with their environ-
ment. Physical computing covers the creative design and realization of interac-
tive objects and installations, which are programmed, tangible media that com-
municate with their environment using sensors and actuators [13]. In physical
computing, many methods and ideas of embedded systems design are used.
2.1 Representation in Curricula, Standards and Recommendations

The ACM CSTA K–12 CS standards [18] range from kindergarten to twelfth
grade and are well-known in and accepted by the international community. Here,
robotics is mentioned as a suitable context for young learners that engages con-
structionist learning. For teaching algorithms in K–6, among other activities,
they suggest to let pupils give algorithmic descriptions to find ways out of a
maze, e.g. using toy robots. From a societal perspective, the impacts of ubiqui-
tous and pervasive computing in daily life are discussed. In later stages, robotics
is a topic when talking about computers as models of intelligent behavior. In the
2016 revision, the creative and design-oriented parts of CS no longer focus on
software development only, but also include the development of physical objects,
e.g. prototypical embedded systems [19]. Internationally, also England’s national
curriculum with the “Computing programmes of study” gained a lot of attention
in recent years. All children from ages 5–16 (key stages 1–4) now have mandatory
CS education in schools [21]. Already in primary school (key stages 1 and 2),
“pupils are equipped to use information technology to create programs, systems
and a range of content” [4]. In key stage 2, specifically, they “design, write and
debug programs that accomplish specific goals, including controlling or simulat-
ing physical systems” [4]. In the accompanying teachers guideline, Computing
At School (CAS) recommends programmable toys (e.g. Bee-Bots, Big Traks) in
key stage 1 and programmable bricks (e.g. LEGO WeDo) in key stage 2 to let
pupils control physical systems and work with sensors, lights and motors.
In Germany, there are recommendations by the German Informatics Society
(GI), the so-called educational standards, which today are used in many federal
states as a basis when revising curricula [5,6]. For the development of computing
systems in class, the GI standards for the upper secondary level recommend
identifying and modeling scenarios from real-world contexts. Typical application
areas that are listed include robotics, process regulation and process control [6,
p. 11]. This development is also evident in recent curricula of some federal states.
For example, the Berlin-Brandenburg CS core curriculum suggests the realization
of a physical computing project and the use of external hardware within the
subject area computing systems [20, p. 23]. In the field of physical computing,
the characterization of embedded systems, the processing of electrical quantities
or the use of microcontrollers in different contexts is called for [20, p. 27].
2.2 Realization in the Classroom

Due to its occurrence in the newer curricula, many practical reports cover experi-
ence in physical computing activities. Often, however, they lack creative aspects
of physical computing and rather fall into the category of robotics. Topics that
were earlier done using LEGO Mindstorms robots (e.g. obstacle avoidance or
line following) are now realized with different hardware, such as Arduino-based
robots (e.g. mBot). The creative potential offered by modern tools, such as
Arduino or Raspberry Pi, is often neglected (e.g. in [2,11]). On the other hand,
there are also many examples from all over the world, where physical computing
is practiced in a creative way—unfortunately, often still in out-of school contexts
only (e.g. [3,8]).
3 Professional Development in Physical Computing

We developed a concept for professional development in physical computing to
qualify CS teachers in very limited time. They should be empowered to not
only implement existing projects, but also to plan their own physical computing
lessons tailored to their particular needs and classroom situations, as described
in [15]. The concept consists of six design principles for professional develop-
ment. Important aspects include networking and follow-up experiences to allow
teachers to share and discuss their ideas and experiences with colleagues and to
offer help with the implementation of pilot projects, support them with hard-
and software issues, give advice when needed and, from a research perspective, to
gain information about their experience. One year after the first two-day train-
ing, we met for an extended workshop where the teachers could connect back
to each other and exchange their experience. We also encouraged inexperienced
teachers to participate, so that they could learn from their peers.
4 Investigating Teachers’ Expectations and Experience

From investigations in our early workshops we found that teachers were strongly
interested in physical computing as a topic for their CS classes. They seemed
exceedingly motivated despite concerns that they might not have enough time or
that technical failure occurred. Interestingly, our impression was that teachers
who had done projects with their classes, mostly did so using new tools to teach
familiar content (e.g. introduction to programming) – it seemed as if physical
computing was used as ‘just another teaching method’ rather than a new content
area. Our experiences from those workshops are described more detailed in [15].
4.1 Aims and Scope of the Study

In the interview study described in this article, the aim was to investigate
teacher’s expectations, beliefs and attitudes in physical computing (in partic-
ular aims, fears, hurdles they see coming in the classroom, mental reservations)
and their relationship to teachers’ classroom experience to better understand

how they can be supported in translating their initial enthusiasm for the subject
into lessons with new learning objects. Our research questions are:
– Why are teachers interested, which benefits do they see?

– Which topics and content do they associate with physical computing?
– Which concerns and challenges do they (fear to) face in the classroom?
– What kind of support (teaching aids, materials, education) do they want?
4.2 Setting and Participants

The workshop was held in Brandenburg, Germany, on two consecutive days.
Our 15 participants (seven female and eight male) came from different German
federal states and Switzerland. Their teaching experience ranged from less than
two years to more than twenty years, with approximately one half of the partic-
ipants having more than ten years of teaching experience. Most teachers came
from different types of secondary schools, two participants were university teach-
ers. Concerning their prior physical computing experience, we had a well-mixed
group. Eight teachers had already held their own lessons, seven of them gained
their basic knowledge in a professional development workshop. The other seven
teachers had no classroom experience. Four of them had until then only worked
independently on the subject or had taken part in a short workshop. Three
teachers had never worked with physical computing tools before, neither in the
classroom, nor in any other context. Thus, through the composition of the group
of participants, we also had the possibility to draw comparisons between teach-
ers with and without practical physical computing experience (in the following
abbreviated to pE-teachers and nE-teachers respectively, Fig. 1).
Fig. 1. Participants’ physical computing experience

4.3 Methodology and Implementation

We conducted semi-structured interviews in order to react flexibly during the
survey with questions of clarification or when new, unforeseen aspects occurred.
Each participant was interviewed once during the workshop, starting with nE-
teachers to avoid that they were influenced by the pE-teachers’ reports. We also
collected final impressions in a group interview at the end of the second day. With
the design of the interview guideline we adhered to Hussy and Mayring [7,10]
and followed a multi-step process from the drafting of a first set of questions,
subsequent discussion of this catalog in our working group, a telephone test inter-
view, the revision and re-discussion of the questions, and eventually the design
of the final question catalog. In total, about 30 questions and sub-questions were
developed and used as interview guide. We evaluated the study using qualita-
tive data analysis methods, as we were interested in the variety of expectations
and experiences. We then clustered our participants based on their experience
to identify trends of change in their attitudes. During the survey design, we also
developed the category system for the data analysis and later, inductively, added
categories for items that were not yet represented in the system.
5 Results and Interpretation

5.1 Interest and Benefits
One aspect of our survey was to investigate, why teachers are interested in
physical computing and which benefits and advantages they find promising for
their teaching. The survey included questions about their general expectations of
physical computing in CS education and, more specifically, expectations regard-
ing the possible effects on their pupils, particularly in relation to motivation and
gender-specific issues. Our hypothesis that teachers perceive didactic rather than
content-related aspects behind physical computing, has been confirmed in the
evaluation of responses to the first question set. In general, the vast majority of
participants considered it an advantage that through physical computing, pupils
got direct feedback and CS became tangible. It is interesting to see the shift
between inexperienced and experienced teachers: some aspects, such as inter-
disciplinary teaching or motivation, were mentioned considerably more often by
nE-teachers. In contrast, practical relevance was more often regarded as a major
benefit of physical computing by the pE-teachers. Additional singular responses
included the creation of tangible products, social skills and creativity. The imbal-
ance between teachers with and without physical computing teaching experience
shows that, e.g. through the experience they have or lack, teachers weigh the
importance of different aspects differently (see Table 1).
5.2 Contents and Topics

When we asked the teachers, which contents they consider relevant for teaching
physical computing, the overall impression was that they connect quite a vari-
ety of topics with it, e.g. programming and algorithms, modeling, sensing and
Table 1. Teachers’ expectations concerning benefits in physical computing depending

on whether they have practical experience (pE, 8 total) or not (nE, 7 total)
General benefits # nE-teachers # pE-teachers Sum

Creativity 1 0 1
Social competence 0 2 2
Interdisciplinarity 3 0 3
Direct feedback and tangiblity 5 6 11
Practical relevance 2 5 7
Product 1 2 3
Motivation 5 1 6
Project goals 1 0 1
actuation, technical aspects of CS or the Internet of things (IoT). However, our

data also shows that in their teaching, teachers mainly connect familiar content
with the new tools physical computing offers. This can be inferred when look-
ing at the differences between the two teacher groups, as depicted in Table 2.
For example, it is noticeable that CS teachers who have already gained practical
experience with physical computing, are more likely to name basic content of CS
such as modeling, programming and algorithms (about 85%). Teachers without
this experience proportionally more often mention specific content that is more
closely linked to physical computing, such as sensing and actuation or the IoT
(about 50%).
Table 2. CS content relevant in physical computing teaching as mentioned by teachers

depending on whether they have practical experience (pE, 8 total) or not (nE, 7 total)
CS content # nE-teachers # pE-teachers Sum

Computer engineering 0 3 3
Embedded systems 0 0 0
Internet of Things 1 0 1
Interactive systems 0 0 0
Sensing and actuation 3 4 7
Measurement, regulation and control 1 0 1
Modeling 0 2 2
Programming and algorithms 4 9 13
Other 1 4 5
sum 10 22 32
In a second question, we examined, to what extent teachers would integrate

the following topics into their lessons: embedded systems, sensing and actuation,
interactive systems and IoT and smart objects. When counting the topics that
teachers consider important and want to talk about in their lessons, interestingly,
sensing and actuation is the only content area, which is mentioned more often by
pE-teachers than by nE-teachers. All other topics are dominated by the other
group. Given the small number of interviewees, these findings should not be
overrated. However, it might be concluded that teachers – although willing to
integrate new topics in the beginning – find it difficult to prepare those contents
for their pupils or have other obstacles that prevent them from doing so. When
looking more closely into their statements1 , it becomes evident that the main
reasons for their hesitance are:
– Lack of knowledge: (e.g. “Embedded system is still like a foreign word for me,
I have no idea what that is.”)
– Topic not in focus: (e.g. “Well, so far, sensing and actuation and interactive
systems. I never thought about the others.”)
– More suitable in different context: (e.g. “The Internet of things, for me, is
more connected to home technologies than physical computing.”)
– Not in curriculum: (e.g. “Well, these are no topics that can directly be found
in the curriculum.”)
As most of the existing curriculum documents are recommendations or

roughly delineated, teachers require more concise descriptions and explanations
of relevant contents and their interrelations. In other subject areas, there are
textbooks and collections of instruction material that have been created over the
years. For physical computing, such aids are not yet available. Thus, teachers
need support in the concrete design and implementation of teaching scenarios.
5.3 Concerns and Challenges
One of the main goals of the survey was to find out what problems CS teachers
expect to occur in physical computing teaching, which challenges they actually
face in the classroom and how they cope with these hurdles. Here, it is particu-
larly interesting to look at differences between pE- and nE-teachers.
Among the general answers to this question, time was seen as a major issue
by the nE-teachers. Among the pE-teachers, this issue was mentioned only once.
However, it is not always clear if teachers fear a lack of time in the classroom
or high preparation times. In the first case, anchoring physical computing in
the curricula might help. In both groups, organizational complexity is seen as a
challenge when planning physical computing projects. pE-teachers often report
how they coped with issues such as finding storage room for the unfinished
projects or receiving funds to buy necessary hardware components. Technical
difficulties were mentioned equally often by both groups and there seems to be
a tendency that pE-teachers experience hardware issues more often than other
technical problems (Table 3).
1
All statements were translated from German by the author.
Table 3. Teachers’ concerns and perceived challenges in physical computing teaching

depending on whether they have practical experience (pE, 8 total) or not (nE, 7 total)
Perceived challenges # nE-teachers # pE-teachers sum

Technical difficulties 4 3 7
Organizational complexity 3 5 8
Unrealistic project goals 0 1 1
Financial difficulties 2 1 3
Time 5 1 6
Too much crafting 1 2 3
None 0 1 1
sum 15 14 29
We posed additional questions for some specific problems that we were con-
fronted with in our previous workshops: When inquired in more detail, nE-
teachers also mentioned several times that they were afraid their pupils might
not be experienced enough to use the necessary hardware tools. Only one of the
pE-teachers confirmed this worry. Many teachers of both experience groups fear
that components of construction kits might break or get lost in class. Not all
pE-teachers support this – pupils, according to one teacher “[...] have a natural
respect for things that are expensive. It’s the same with a laptop, they wouldn’t
throw it on the floor.” However, there was a tendency that teachers do not regard
this a problem in Gymnasium2 , but rather in other secondary school types that
do not focus on academic learning (Fig. 2). Other challenges were only mentioned
once or twice and seemed to be of minor importance (e.g. unrealistic project goals
or too much crafting in the classroom).
no pract. exp.
13% 13%
pract. exp.
no problem
Gymnasium
small problem
University 27%
medium problem
other sec. school 47%
large problem
0 2 4 6
no or small problem medium or large problem
Fig. 2. Teachers’ concerns that students loose or break parts of construction kits
2
Gymnasium: advanced secondary school that leads to Abitur (A-Level equivalent),
compares to grammar schools (UK) or preparatory high schools (US).
Table 4. Support strategies and means for teachers depending on whether they have
practical experience (pE, 8 total) or not (nE, 7 total)
Category Code # nE-teachers # pE-teachers sum

Strategies collaboration 4 0 4
professional development 1 0 1
networking 2 0 2
detailed planning 2 4 6
other 3 1 4
Means of support financial support 2 1 3
best practice examples 5 2 7
lesson plans 2 4 6
textbooks for teaching 3 1 4
other 2 4 6
5.4 Support
We also tried to figure out how teachers can be supported in avoiding prob-
lems in the future. They gave different answers depending on the particular
problems. While nE-teachers strive for networking opportunities, professional
development and collaboration both with colleagues and external partners, in
contrast, pE-teachers also emphasize the need for detailed descriptions and man-
uals and elaborate lesson plans (Table 4). Well-developed, tested and reliable
tools seem to be really important for successful teaching. If teachers have to deal
with technical difficulties, this is often perceived as problematic. However, there
were also many of the teachers’ initial concerns that were not confirmed by the
pE-teachers. Thus, networking meetings, where experiences are exchanged and
experts are available for questions are important, especially for inexperienced
teachers. Guidelines and teaching aids can help during the implementation phase,
scaffolding the process of physical computing lesson planning. In the last ques-
tion, we asked teachers what they would find helpful with the implementation
of physical computing projects. In general, the answers reflect the findings from
the previous sections pretty well: example projects and lesson plans are men-
tioned most frequently, directly followed by text books. Financial support and
networking were mentioned twice, each. Additional aspects were, for example,
classroom-suitable construction kits, more workshops and professional develop-
ment opportunities, tutorial videos and ideas how to deal with grading (category
“other” in Table 4).
6 Conclusions
In this study we analyzed teachers expectations, attitudes and experience with
physical computing to gain a better understanding of their aims and needs in
teaching. Our key findings are:
1. Interest in the subject mainly stems from pedagogical considerations: Benefits

of physical computing are seen in motivation, direct feedback and tangibility
of CS and not primarily in new contents related to the field. This is, partly
at least, due to the fact that many publications focus on the aforementioned
aspects only.
2. Related to the first finding, teachers mainly associate traditional topics of CS
with physical computing: pE-teachers reported they were using physical com-
puting tools to teach in content areas such as algorithms and programming
and not, as one might expect, topics such as embedded systems or IoT. The
reasons given suggest that more professional development opportunities are
required in the new subject areas and that these topics must also be anchored
in curricula.
3. Participants suspected and experienced a high organizational effort: nE-
teachers saw most of the challenges in the acquisition of hardware and mate-
rials, the greater complexity of the necessary preparations, and the overall
effort involved in the development and adoption of new processes and rou-
tines. The pE-teachers added challenges in project management, organization
of the classroom, as well as the storage of the components and unfinished
projects. There is a need for practical solutions to these problem areas, in
particular to reduce the initial barriers for hesitant teachers.
4. Teachers need support in many areas: above all, they call for example projects
that give orientation, but also the development of suitable tools, classroom
materials and manuals are required, as well as networking opportunities.
5. Interdisciplinary teaching is hardly practiced: collaborations are primarily
pursued with art and physics teachers, often combined with the aim of out-
sourcing the craft work on the projects from computer science classes. Again,
suggestions and practical examples are required.
Some work has already been done in all of these sectors. There are tools
available on the market, some of them more and some of them less suitable
for education (cf. [14]), many teachers piloted physical computing projects and
developed materials. In general, however, more needs to be done in all the men-
tioned areas: curricula need revision, tools need improvement and professional
development is required that goes beyond an introductory course on tool oper-
ation and especially focuses on deeper examinations of the new content. Only
then we can provide successful education in the field of embedded systems, which
enables our pupils to appropriately participate in the digital society.
References
1. BITKOM: Eingebettete Systeme – Ein strategisches Wachstumsfeld für Deutsch-
land [Embedded Systems – A stratgic growth area for Germany] (2010).
https://www.bitkom.org/noindex/Publikationen/2010/Leitfaden/Eingebettete-Sy
steme-Anwendungsbeispiele-Zahlen-und-Trends/EingebetteteSysteme-web.pdf.
Accessed 07 Sept 2017
2. Brinkmeier, M., Kalbreyer, D.: A case study of physical computing in computer

science education. In: Vahrenhold, J., Barendsen, E. (eds.) Proceedings WiPSCE
2016, pp. 54–59. ACM, New York (2016)
3. Buechley, L., Eisenberg, M., Elumeze, N.: Towards a curriculum for electronic
textiles in the high school classroom. ACM SIGCSE Bull 39(3), 28 (2007)
4. Department for Education: Computing programmes of study: key stages 1 and 2.
National curriculum in England (2013)
5. Gesellschaft für Informatik e.V.: Grundsätze und Standards für die Informatik
in der Schule [Principles and standards for computer science at school]. LOG IN
28(150/151), supplement (2008)
6. Gesellschaft für Informatik e.V.: Bildungsstandards Informatik für die Sekun-
darstufe II [Educational standards for computer science in upper secondary edu-
cation]. LOG IN 36(183/184), supplement (2016)
7. Hussy, W., Schreier, M., Echterhoff, G.: Forschungsmethoden in Psychologie und
Sozialwissenschaften für Bachelor. Springer, Berlin Heidelberg (2010). https://doi.
org/10.1007/978-3-642-34362-9
8. Kafai, Y.B., Lee, E., Searle, K., Fields, D.: A crafts-oriented approach to computing
in high school: introducing computational concepts, practices, and perspectives
with electronic textiles. ACM T Comput. Ed. 14(1), 1–20 (2014)
9. Lee, E.A., Seshia, S.A.: Introduction to Embedded Systems - A Cyber-Physical
Systems Approach, 2nd edn. MIT Press, Cambridge (2017)
10. Mayring, P.: Qualitative Content Analysis. Theoretical Foundation, Basic Proce-
dures and Software Solution. Klagenfurt, Austria (2014). http://nbn-resolving.de/
urn:nbn:de:0168-ssoar-395173
11. Neutens, T., Wyffels, F.: Teacher professional development through a physical
computing workshop. In: Proceedings WiPSCE 2016, pp. 108–109, no. 6 (2016)
12. Przybylla, M., Israel, P., Streichert, J., Romeike, R.: Bridging motivation gaps with
physical computing in CS education. In: ISSEP 2016, p. 53 (2016)
13. Przybylla, M., Romeike, R.: My interactive garden-a constructionist approach to
creative learning with interactive installations in computing education. In: Kynigos,
C., Clayson, J.E., Yiannoutsou, N. (eds.) Constructionism: Theory, Practice and
Impact. Proceedings of Constructionism 2012, pp. 395–404 (2012)
14. Przybylla, M., Romeike, R.: Key competences with physical computing. In: Pro-
ceedings of Key Competencies in Informatics and ICT (KEYCIT) 2014. Univer-
sitätsverlag Potsdam, Potsdam (2014)
15. Przybylla, M., Romeike, R.: Teaching computer science teachers - a construction-
ist approach to professional development on physical computing. In: Sipitakiat,
A., Tutiyaphuengprasert, N. (eds.) Proceedings of Constructionism 2016 (Con-
structionism 2016, Bangkok, Thailand), pp. 265–274. Suksapattana Foundation
(2016)
16. Przybylla, M., Romeike, R.: Social demands in ubiquitous computing : contexts
for tomorrow’s learning. In: Webb, M., Tatnall, A. (eds.) 11th World Conference
on Computers in Education. Springer International Publishing, Dublin (2017)
17. Richards, M., Petre, M., Bandara, A.K.: Starting with UbiComp: using the sense-
boards to introduce computing. In: 43rd ACM Technical Symposium on Computer
Science Education, pp. 583–588. ACM, Raleigh (2012)
18. Seehorn, D., Carey, S., Fuschetto, B., Lee, I., Moix, D., O’Grady-Cunniff, D.,
Owens, B.B., Stephenson, C., Verno, A.: CSTA K–12 Computer Science Standards.
Technical report, New York (2011)
19. Seehorn, D., Primann, T., Lash, T., Lee, I., Moix, D.: [INTERIM] CSTA K-12
Computer Science Standards. Technical report, New York (2016)
20. SenBJW Berlin, MBJS Brandenburg: Teil C Informatik Wahlpichtfach [Part C

Computer Science Elective Course] (2015)
21. Sentance, S., Humphreys, S.: Online vs face-to-face engagement of comput-
ing teachers for their professional development needs. In: Brodnik, A., Varen-
hold, J. (eds.) Informatics in Schools. Curricula, Competences, and Competi-
tions, vol. 9378, pp. 69–81. Springer, Cham (2015). https://doi.org/10.1007/
978-3-319-25396-1 7
22. Vahid, F., Givargis, T.D.: Embedded System Design: A Unified Hardware/Software
Introduction. Wiley, New York (2002)
23. Weiser, M.: Some computer science issues in ubiquitous computing. Commun. ACM
36(7), 75–84 (1993)
Teacher Development and Assessment
Investigating Informatics Teachers’ Initial
Pedagogical Content Knowledge on Modeling
and Simulation
Nataša Grgurina1(&), Erik Barendsen2,3, Cor Suhre1,

Klaas van Veen1, and Bert Zwaneveld3
1
Teaching and Teacher Education, University of Groningen,
Groningen, The Netherlands
{n.grgurina,c.j.m.suhre,klaas.van.veen}@rug.nl
2
Radboud University, Nijmegen, The Netherlands
e.barendsen@cs.ru.nl
3
Open University, Heerlen, The Netherlands
zwane013@planet.nl
Abstract. Computational science, comprised of modeling and simulation, is a

new theme in the new 2019 Dutch secondary education informatics curriculum.
To investigate the pedagogical content knowledge (PCK) on modeling and
simulation, we interviewed ten informatics teachers and analyzed their PCK,
distinguishing its four elements - knowledge of goals and objectives, students’
understanding, instructional strategies and assessment - and investigated
potential differential features of their PCK in order to typify teachers’ individual
PCK. We charted the teachers’ PCK in terms of these four elements and found
differential features related to knowledge of goals and objectives and related to
knowledge of assessment, dividing these teachers into four distinct groups.
However, these differential features do not lead to distinct types of PCK. Our
findings will be used to explore the future development of teachers’ PCK and
they will contribute to the development of teaching materials, assessment
instruments and teacher training courses on modeling.
Keywords: Pedagogical content knowledge Computational science

Modeling and simulation Informatics Secondary education
1 Introduction
Modeling plays a significant role in the development and learning of science [13] and
informatics provides the means for students to actively engage in learning science by
providing tools and techniques to engage in modeling. The new 2019 Dutch secondary
education informatics curriculum recognizes this and includes an elective theme
comprised of modeling and simulation, together called Computational Science. It is
described by the high-level learning objectives: “Modeling: The candidate is able to
model aspects of a different scientific discipline in computational terms” and “Simu-
lation: The candidate is able to construct models and simulations, and use these for the
research of phenomena in that other science field.” Modeling itself will be a part of the

https://doi.org/10.1007/978-3-319-71483-7_6
66 N. Grgurina et al.
compulsory core curriculum, described as “Modeling: The candidate is able to use

context to analyze a relevant problem, limit this to a manageable problem, translate
this into a model, generate and interpret model results, and test and assess the model.
The candidate is able to use consistent reasoning” [2, 8]. The curriculum does not
provide further details about these objectives, instruction or assessment. In line with the
Dutch tradition, this is left to educators and authors of teaching materials. The elab-
oration of this learning objective, the development of teaching materials, assessment
tools and teacher training courses are already taking place and we intend to monitor
these developments.
This study is a part of a larger research project on teaching Computational Science
in the context of informatics in Dutch secondary education, investigating pedagogical
aspects and teachers’ pedagogical content knowledge (PCK) about modeling1. Fol-
lowing Magnusson et al. [15], we distinguish four elements of content-specific peda-
gogy: (M1) goals and objectives, (M2) students’ understanding and difficulties, (M3)
instructional strategies, and (M4) assessment. Previously, we refined the CSTA defi-
nition of computational thinking (CT) [5], made initial explorations of teachers’ PCK
[6, 7] and of the computational modeling process [9], obtained an operational
description of the intended learning outcomes of the learning objective Computational
science – thus focusing on Magnusson’s element M1, observed students working on
modeling tasks - focusing on Magnusson’s element M2, and established what data
sources were suitable for assessment - Magnusson’s element M4 [8].
Aim of this study. In this study, we focus on teachers’ PCK on modeling. We address
the following research questions:
1. How can the teachers’ PCK be portrayed in terms of the four elements of PCK?
2. What differential features of PCK can be used to typify teachers’ individual PCK in
terms of the four elements of PCK?
Our findings will serve as input for the subsequent studies on designing teaching
materials and assessment instruments, the development of teachers’ PCK, and they will
contribute to the development of teacher training courses.
Related Work. The construct of PCK has proven to be a powerful one to help capture
teachers’ views and knowledge on teaching various topics in STEM disciplines such as
models and modeling in science [13] and as a part of public understanding of science
[11], programing in informatics [17] and designing digital artifacts in informatics [16];
to expose the relation between the quality of teachers’ PCK and their subject matter
knowledge [17, 18] to explore the PCK repertoire of beginning teachers [14] and to
chart the development of teacher’s PCK as the experience with teaching a particular
topics increases [10].
We investigate PCK through the lens of the operational description of the intended
learning outcomes of the learning objective Computational science obtained in our
previous study, describing the modeling cycle for simulation modeling through its
1
In this paper, the terms modeling, simulation modeling and computational science all refer to the
learning objective computational science.
Investigating Informatics Teachers’ Initial Pedagogical Content Knowledge 67
elements purpose, research, abstraction, formulation, requirements/specification,

implementation, verification/validation, experiment, analysis, and reflection [8].
Context of the study. In the Netherlands, informatics is an elective subject in grades
10 and 11 of the senior general secondary education spanning grades 7 through 11 (in
Dutch: HAVO) and in grades 10 through 12 of the pre-university education spanning
grades 7 through 12 (in Dutch: VWO). In grades 10 through 12, all students follow
either one of the two humanities tracks or one of the two science tracks and they can all
elect to follow informatics if their school offers it. There are two routes to teachers’
qualification for informatics. When informatics was introduced in secondary education,
in the early 2000’s, teachers of other subjects could get qualified through in-service
training. As of 2006, qualification entails an MSc degree in informatics education [4].
2 Method
We conducted individual semi-structured interviews with ten informatics teachers from

the local informatics teachers’ network who replied to our invitation to be interviewed.
Four of the teachers have an informatics background and three of these are qualified
teachers with an MSc degree in informatics education while the fourth one is still
studying to get this qualification. Out of the other six teachers, one has no teacher
qualification, one is qualified for mathematics only, and other four are qualified
teachers for other subjects - such as physics or history - who additionally attained
teacher qualification for informatics through in-service training. The teaching experi-
ence of these ten teachers ranges from a few months to several decades. Eight of these
teachers were about to take a course on agent-based modeling. Interviews lasted
between half an hour and an hour, depending on the extent of a particular teacher’s
PCK. In our interview protocol, we first enquired about the teachers’ educational and
professional background and whether they had already taught modeling. We then asked
detailed question arranged around the four PCK elements described by Magnusson,
that were inspired by [11–13, 16].
• M1: Goals and objectives
What comes to mind when you hear the word ‘model’? In which context(s) does
this word make sense to you? [12] What are models for? In which circumstances are
they used? [12] What do you expect to be your main objective in teaching modeling
and simulation in informatics? [11] Why do you intend to teach this to your students
in informatics? What do expect you will like or dislike about modeling projects by
your students? [16]
• M2: Students’ understanding
Do your students need any specific prerequisite knowledge to be able to learn about
modeling and simulation in informatics? [11] What sorts of skills do students need
to acquire in order to be able to develop models and run simulations? [16] What do
you expect to be successful for your students? [11, 16] What do you expect to be
difficulties for your students? [11, 16] What do expect your students to actually
learn from their modeling (and simulation) projects? [16]
• M3: Instructional strategies

In what activities and in what sequence do you expect your students to participate in
the activities of learning modeling and simulation? What to teach students to achieve
the modeling objectives? [16] How to teach students to achieve the modeling
objectives? [16] What do you expect to be your role as a teacher when teaching about
modeling and simulation? [11] What do you expect are going to be the teaching
difficulties/problems concerned with the modeling projects in your classroom? [16]
What technological tools do you intend to use in your classroom? [16]
• M4: Assessment
How do you intend to assess your students’ learning and achievement during their
modeling projects? [16] How do you intend to establish whether your students
reached the learning goals with regard to modeling and simulation? How would you
know? [11]
The interviews were recorded and transcribed verbatim. We first coded the tran-
scripts using the coding categories derived from our operational description of mod-
eling [8]. We then classified the interview transcripts using Magnusson’s four elements
of PCK [15] as main coding categories. Within these categories, we applied inductive
coding to characterize the teachers’ responses. In an axial coding process [3], the codes
were grouped and merged where necessary. We used the codes to describe the teachers’
PCK. In a subsequent analysis, we tried to identify differential features with the pur-
pose of typifying teachers’ individual PCK.
3 Results
In this section, we first present the results of our characterization of PCK organized
around the four elements of PCK. We then explore differential features in order to
distinguish types of teachers’ PCK.
3.1 Knowledge About Goals and Objectives (M1)

No teacher has taught computational science as a separate topic in the context of the
informatics course yet and only one of them taught system dynamics modeling in a
physics course. Since we did not enquire about the modeling process explicitly, we
performed a detailed analysis of the interviews through the lens of our theoretical
framework on the modeling cycle and thus reconstructed the teachers’ content
knowledge (CK) pertaining to the aspects of modeling, as shown in Table 1.
Concerning the nature of models, six teachers said that models were a simplified
representation of reality, three reported that models were something to be used for
calculations, two saw models as something for visualization and communication, and
one also mentioned physical models. Regarding the contexts for the use of models, the
teachers mentioned information systems, scientific research and cited several examples
(e.g. exploring group forming on ethnical basis). Several teachers stressed the fact that
informatics served other disciplines - an idea that persistently permeates their thinking
about modeling and teaching modeling.
Table 1. Teachers CK on modeling cycle

Aspect Teacher
1 2 3 4 5 6 7 8 9 10
Purpose ✔ ✔ ✔ ✔ ✔ ✔
Research ✔
Abstraction ✔ ✔ ✔
Formulating ✔ ✔ ✔ ✔ ✔ ✔
Requirements/specification ✔ ✔
Implementation ✔
✔ ✔ ✔
Verification/validation ✔ ✔ ✔ ✔ ✔ ✔ ✔
Experiment ✔ ✔ ✔ ✔ ✔
Analysis ✔ ✔ ✔ ✔
Reflection ✔ ✔ ✔ ✔ ✔ ✔ ✔
According to the teachers, the objectives of teaching modeling are twofold:

• Conceptual objectives: these objectives emphasize learning to master skills asso-
ciated with informatics subject matter. Teachers mention CT aspect automation,
software design cycle, the necessary research skills, analysis of the world the stu-
dents live in, linking and translating reality to model through abstraction, building a
model and, in words of teacher 9, “using it […] so that you can predict things or test
things, or … ehm …, understand things”.
• Motivational and practical objectives: these objectives focus on students’
engagement, motivation, attitude, skills, practical benefits, insight and awareness
about relevance. Teachers mention perseverance, building confidence about stu-
dents’ own ability, developing self-reliance, and realizing that models are useful
tools that can be used in specific situations. In this light, learning to model helps to
enhance students’ insight, it helps develop cooperation and communication skills, it
is interesting or fun, it serves as preparation for the future, and it concerns “soft
informatics”. Furthermore, modeling lays a connection between informatics and
other disciplines and models could be useful for students’ subsequent studies. In
words of teacher 10, “… that you need to be able to design a test setup with
informatics, ehm, digital means so that you can get a lot of results, but processing
these results is no more informatics, that’s physics.”
3.2 Knowledge of Students’ Understanding (M2)

The prerequisite knowledge needed to learn modeling and skills needed to make models
are related to:
• Student’s characteristics: such as age and development, mindset, attitude, and
other skills. For example, teacher 6 believes “it is all very complex, they [students]
are only 15 years old” and teacher 7 is reminded of Bloom’s taxonomy and fills in
“it gets easier as they get older”. Fantasy, creativity and an analytical mindset also
play a role. Attitude is important: several teachers stress the significance of perse-
verance. Some mention differences among students: teacher 6 expects to see dif-
ferent skill levels when it comes to tackling practical assignments.
• Student’s knowledge pertaining: the other discipline, programming or computa-

tional thinking, and modeling aspects. Teachers 4, 5, 6 and 8 expect the students to
know something about the phenomenon from other discipline they are modeling.
Almost all the teachers expect their students to be familiar with programming before
they embark on modeling. However, teacher 9 also sees the possibility to use
modeling as a vehicle to teach programming. Teachers 4, 8 and 10 expect their
students to be familiar with several modeling aspect such as being able to explain
how a particular model works, abstraction, and in words of teacher 8, “you need to
learn to recognize the actors and you need to be able to see the relations among
them, to …ehm … to describe them, to translate them into a model”.
• The chosen teaching approach: teacher 8 mentions the interplay between the
teaching materials used and the necessary prerequisite knowledge and says “but you
need to sense what extras you should offer” and adds, “It’s quite abstract so I think
about the 12th grade, but if you present it simply enough, then you could teach it in the
10th grade as well, without ever having taught any informatics before that”. He also
stresses the importance of re-activating the knowledge the students already possess.
The issues teachers regard as successful:
• Relevance: through modeling, learning about familiar phenomena encountered in
other courses (e.g. biology, economy) rather than distant ideas. As teacher 6 puts it:
“You concoct nice stories about nuclear power plants, but how many children end
up in there?”
• Perception: attitude, experiencing success, interest and fun, and student’s charac-
teristics. Several teachers expect students’ confidence would grow through their
perseverance when facing problems and experiencing success by making a model
on their own. Teacher 2 admits that achieving this with his students poses a chal-
lenge. Numerous teachers expect these aspects, together with the relevance expe-
rienced by students and the possibility to come up with creative solutions and
implementation of students’ own ideas, to make modeling interesting and fun.
Teacher 10 mentions Gardner’s multiple intelligences to explain why he expects
students in science tracks to perform differently than students in humanities tracks.
• Skills: programing and computational thinking. Most teachers expect that pro-
gramming the models would not pose a problem.
• Organizational issues: teaching strategies to meet students’ needs. Several teachers
describe how to contribute to the students’ success by choosing a suitable teaching
approach. A number of teachers who intend to use NetLogo to teach modeling prize
its user-friendliness and see this as a success factor.
• Interest and fun: Teacher 5 says, “even if I never teach this, I still find it fun for
myself” and goes on to reflect how to transfer this enthusiasm to his students.
• Finally, teachers 1, 3, 4 and 9 do not know what to say. As teacher 4 puts it, “I’ve
never taught modeling, so I wouldn’t know what would be a success.”
The issues teachers regard as difficult:
• Technical issues: programing and computational thinking, aspects of modeling and
development. For example, the meaning of the term parameter in physics or
mathematics differs from the meaning in a simulation model and that might be
confusing. Students who follow physics course are already familiar with models, as
opposed to students in humanities tracks. Teacher 2 expects that in the beginning,
students will have difficulties understanding existing models and how their com-
ponents interact. Teacher 4 expects her students to have problems getting used to
the programming language and subsequently to have problems programming.
Numerous teachers expect that modeling aspects such as abstraction – deciding
what is relevant for a model and what to leave out - will be difficult. Some teachers
expect problems with implementation – translating conceptual model into program
code. Teacher 7 expects his students could be too ambitious with the models they
want to make and suggests keeping an eye on his students all the time to be able to
intervene and help in case they encounter any of these problems.
• Perception: attitude, skills, interest or fun, relevancy, age and development. Tea-
cher 9 expects problems with motivation if the students do not see the relevance of
modeling. He also mentions lack of perseverance and inability to go on after getting
stuck. Teacher 6 believes that abstract aspects of modeling are difficult for the
students of this age – 16 years old. He also mentions students not using common
sense.
• Approach: work method, possibility to work on their own case. Some teachers
expect problems with students who dive right into building their models without
giving it sufficient thought first, and, with students who lack oversight and do not
know where to begin. According to teacher 9, the last problem could be alleviated
by good teaching material. Several teachers believe some students would have
difficulties coming up with a suitable case to model.
• Again, some teachers do not know what to say.
3.3 Knowledge About Instructional Strategies (M3)

When asked about their teaching approach, some teachers are cautious about replaying
due to lack of experience, but in the end all the teachers have similar ideas that can be
summarized as follows: during a period ranging from six to twelve weeks, scaffold
learning by beginning with an introduction about models in general, show and explain
the working of few models and their code, then give students several assignments to
expand existing models or develop simple models from scratch, possibly differentiate
to account for students interests, level (HAVO or VWO) or the track they follow
(humanities or science), and finally, engage them in a large (group) project where they
develop a model from scratch during several weeks. While describing their teaching
approach with various degree of detail, the teachers report about:
• Their role as a teacher: to coach, to explain, to show, to help, to encourage, to
keep an eye on students’ progress, and simply to be there. Teacher 9 says, “and
then you come in there to steer. And if it gets stuck, not just to answer the question,
but to help them find the answer themselves”. Teacher 5 is cautious: “well, when
they’re working on a new model, I don’t always have a ready-made answer.”
Teacher 7 would have his students recreate a small program to check their under-
standing before they embark on the large project. During the project, he would have
a double role, both as the teacher and as a client. As the teacher, he would keep an
eye on the progress of the whole project. As the client, he would come in every two
weeks and tell his students things like “hey, you have things that make no sense”
and have students fix the problems themselves.
• Assignments the students work on: open and closed problems and making
models. Teacher 6 would have a number of closed problems together with answers
for students to practice before embarking on the open problem they have to work on
as their final project. For the final project, most teachers would let their students
come up with their problems themselves, but would also have a list of problems
available for those who cannot think of something themselves.
• Student’s characteristic to take into account: level and the track they follow.
Teacher 8 would have different assignments for students to practice on, catering to
their needs and preferences, depending on the educational track they follow. Tea-
cher 2 would not require his HAVO students to develop a model from scratch, but
would rather have them expand and adjust an existing model.
• Organizational aspects: the daily teaching practice, planning, organizing, SCRUM,
rapid prototyping and playing the role of the client. Teacher 3 teaches 60-minutes
lessons and would start each lesson with a short central instruction and let the students
work for themselves the rest of the time. He would use Trello boards for planning and,
like teacher 9, employ SCRUM to organize the work. Teacher 10 would send his
students to a teacher of a different subject who would then pose as a client, while he
himself would be a process supervisor. Teacher 9 would pose as a client himself.
• Difficulties and problems: technical problems, problems related to teaching
materials and other problems. Teacher 9 warns that no matter how simple the
software used is, there is always a possibility of getting an error message, not
understanding it and then getting stuck.
3.4 Knowledge About Assessment (M4)

When asked about assessment, all teachers agree that the large practical assignment the
students worked on to learn modeling could serve for the assessment purpose as well.
Teacher 9 could possibly give his students a written exam instead. Teacher 8 would use
a small written exam to ascertain the students learned enough about modeling before
they are allowed to start working on a large project. When talking about assessment,
the teachers report on:
• Assessment form: written exam – formative and summative, project to be done
individually or in groups.
• Problems to work on: given by the teacher or provided by the students themselves,
open or closed, opportunity for differentiation.
• Organizational issues: SCRUM and rapid development.
• Assessing the work: quality of the end product and project documentation, aspects
of modeling cycle, teacher’s impression about the students’ activities in the lessons
during work on the project. When assessing the results of the final project, most
teachers want to look into the quality of the model - and some of them the code too -
as described by the students in the project documentation they are required to turn
in. The teachers mention various aspects of the modeling process as relevant for the
assessment. However, no clear quality criteria are elaborated and teacher 1 would
simply estimate the quality of the project. Teacher 8 adds that technical aspects are
easy to assess, but it is difficult to see if the students realize the full spectrum of
possibilities the modeling offers to them. Some teachers would rely mainly on their
observation of students while they work on the projects, again without elaborating
on specific quality criteria. Additionally, teacher 10 would talk to his students to
ascertain whether they understand what they are doing. Teacher 6 would also assess
the quality of the report the students wrote for the customer. Teacher 7 would have
his students present their models, he would assess the product (i.e. model and
project documentation) and additionally the process (SCRUM) and he stresses the
importance of reflection, together with number of other teachers. He adds, “even if
they didn’t succeed, I find you can’t say they don’t know what modeling is”.
Teacher 5 is not sure what to say: “if the model works, is that sufficient?”
3.5 Differential Features and Typification

We found two characteristics that distinguish among teachers:
• Knowledge of goals and objectives (M1). Teachers 1, 2, 4, and 6 stress the
importance of conceptual objectives such as learning how to employ programming
or how to make models, while teachers 3, 5, 7, 8, 9 and 10 put more emphasis on
broader motivational and practical objectives such as enhancing insight or
preparing for the future.
• Knowledge of assessment (M4). Teachers 1, 2, 4, 8, 9 and 10 predominantly put to
use product-based assessment stressing the quality of the students’ product, while
teachers 3, 5, 6 and 7 prefer process-based assessment stressing the importance of
the employed working procedures.
Combining these two differential features leads to four groups of teachers as shown
in Table 2:
Table 2. Distinct groups of teachers

M1
Conceptual objectives Practical and motivational objectives
M4 Product-based 1, 2, 4, 8, 9, 10
Process-based 6 3, 5, 7
Our analysis revealed, however, that the knowledge of students’ understanding (M2)
and instructional strategies (M3) varies within each of these four groups, so the above
differential features do not give rise to a typification of the teachers’ overall PCK.
4 Conclusion
In answering the first research question - How can the teachers’ PCK be portrayed in
terms of the four elements of PCK? - we portrayed each of these elements:
Concerning teachers’ knowledge about goals of objectives on teaching modeling,

we charted teachers’ content knowledge and described learning objectives in terms of
conceptual, and motivational and practical objectives.
Concerning the knowledge about students’ understanding, the teachers reported
on the prerequisite knowledge, attitudes, skills, abilities and various approaches stu-
dents have to learning, in line with Magnusson et al. [15]. However, when talking
about the difficulties, they mentioned problems due to the abstract nature of modeling
and inefficient students’ strategies, but no teacher mentioned misconceptions.
Concerning the knowledge about instructional strategies, we see an agreement
about subject-specific strategies [15] – scaffolding learning with a final project which
serves both to give students the opportunity to learn how to develop a model from
scratch and as assessment.
Concerning knowledge of assessment, there is an agreement about a suitable form
– a large practical assignment. Teachers mentioned a range of assessment criteria
focused on the quality of students’ products and teachers’ impressions of the students
work process. In contrast to the uniformity regarding the form of assessment, there is a
great variation in granularity and depth of their description of assessment criteria. We
saw similar diversity regarding the knowledge of dimension to assess [15], i.e. the
aspects of modeling process.
In answering the second research question - What differential features of PCK can
be used to typify teachers’ individual PCK in terms of the four elements of PCK? - we
found two characteristics that distinguish among teachers: their focus on conceptual
versus motivational and practical objectives (M1) and their emphasis on product-based
versus process-based assessment (M4) leading to four distinct groups of teachers.
However, none of these differential features leads to an overall typification of the
teachers’ PCK.
5 Discussion
Reflections on findings. As a possible explanation for the variations found in the

teachers’ PCK, we explore the relation with the teachers’ background. Not surprisingly,
we saw that teachers 1, 2, 4 and 9, who all have a background in informatics, displayed
rich knowledge of modeling. Teachers 1 and 4 - both young teachers - had limited idea
about what to expect from their students beyond the general remarks about students
needing to plan before acting and lack of perseverance. On the other hand, teachers 2
and 9 exhibited rich knowledge of students’ understanding and related their extensive
knowledge of instructional strategies to it. Teachers 6, 7, 8, and 10 possess rich and
well-connected PCK. Finally, we saw that teachers 3 and 5, despite their limited
content knowledge, were able to relate their general knowledge of their students to their
teaching strategies. The knowledge of their students’ understanding and difficulties was
the strongest component of their PCK. Other components of their PCK were weaker
and so were the relations among these components too.
Regarding the instructional strategies, the teaching approach described here is in line
with prevailing informatics teaching practices in the Netherlands. The extent of
knowledge of topic-specific strategies [15] varies across teachers and seems to be
related to their subject matter knowledge and teaching experience. The findings about
young teachers 1 and 4 are in line with the results by Lee et al. [14] who found that “a
strong science background does not guarantee a proficient level of PCK.” The more
experienced teachers sometimes behave like novices too, – e.g. teacher 5 – while in
others, their extensive PCK seems to sustain them in the non-familiar area of modeling,
in line with results of Borko et al. [18] who found in a similar situation that rich PCK
for general science topic seems to sustain teachers “in whatever content they are
teaching”.
We observed two characteristics allowing us to distinguish among teachers that
were observed by Rahimi et al. [16] as well. However, our findings are not completely
in line with these findings because in our case, for example, teachers 1 and 10, both
leaning toward predominantly product-based assessment, require students to keep
logbooks to document the modeling problems, difficulties and dead ends they
encountered. On the other hand, teacher 6 would take customer’s feedback into account
and teacher 7 would have his students present their work in the class, while they both
lean toward predominantly process-based assessment. Unlike Rahimi et al., we were
not able to typify teachers’ PCK through relating their knowledge of students’
understanding and instructional strategies on one hand, to their knowledge of goals and
objectives and knowledge of assessment on the other.
Remarkably, despite the great variation of assessment criteria mentioned, there is
hardly any evidence of quality indicators used to judge to what extent these criteria are
met and to what extent the students are able to apply the elements of modeling to a
satisfactory degree.
Limitations of the study. In this study, we charted PCK of a small group of infor-
matics teachers. However, because of the variations in their educational background,
teacher qualification and teaching experience we expect that our findings are fairly
typical for the population of Dutch informatics teachers. This is to be confirmed in
further research.
Implications for educational development. We believe that teachers would benefit
not only from a course on modeling, but also from the availability of teaching mate-
rials. We are convinced that the quality of assessment - an issue attracting a lot of
attention in modern informatics education [1] – would improve if teachers get assis-
tance with designing assessment instruments that would take into account both product
and process.
In the subsequent phases of this research project we will focus on development of
teaching materials and assessment instruments. In parallel, in-service teacher training
based on these findings will be offered to interested teachers. Finally, all the partici-
pants in this study will be followed to chart the development of their PCK on modeling
in the future.
Acknowledgments. This work is supported by the The Netherlands Organisation for Scientific
Research grant nr. 023.002.138.
References
1. Alturki, R.A.: Measuring and improving student performance in an introductory program-
ming course. Inform. Educ. Int. J. 15(2), 183–204 (2016)
2. Barendsen, E., Tolboom, J.: Advisory Report (Intended) Curriculum for Informatics for
Upper Secondary Education. SLO, Enschede (2016)
3. Cohen, L., Manion, L., Morrison, K.R.B.: Research Methods in Education, 6th edn.
Routledge, New York (2007)
4. Grgurina, N., Tolboom, J.: The first decade of informatics in dutch high schools. Inform.
Educ. 7(1), 55–74 (2008)
5. Grgurina, N.: Computational thinking in dutch secondary education (2013)
6. Grgurina, N., Barendsen, E., Zwaneveld, B., van Veen, K., Stoker, I.: Computational
thinking skills in dutch secondary education: exploring pedagogical content knowledge.
ACM (2014)
7. Grgurina, N., Barendsen, E., Zwaneveld, B., van Veen, K., Stoker, I.: Computational
thinking skills in dutch secondary education: exploring teacher’s perspective. ACM (2014)
8. Grgurina, N., Barendsen, E., Zwaneveld, B., van Veen, K., Suhre, C.: Defining and
observing modeling and simulation in informatics. In: Brodnik, A., Tort, F. (eds.) ISSEP
2016. LNCS, vol. 9973, pp. 130–141. Springer, Cham (2016). https://doi.org/10.1007/978-
3-319-46747-4_11
9. Grgurina, N., Barendsen, E., van Veen, K., Suhre, C., Zwaneveld, B.: Exploring students’
computational thinking skills in modeling and simulation projects: a pilot study. ACM
(2015)
10. Henze, I., van Driel, J.H., Verloop, N.: Development of experienced science teachers’
pedagogical content knowledge of models of the Solar System and the Universe. Int. J. Sci.
Educ. 30(10), 1321–1342 (2008)
11. Henze, I., van Driel, J.H., Verloop, N.: Science teachers’ knowledge about teaching models
and modelling in the context of a new syllabus on public understanding of science. Res. Sci.
Educ. 37(2), 99–122 (2007)
12. Justi, R., Gilbert, J.: Teachers’ views on the nature of models. Int. J. Sci. Educ. 25(11),
1369–1386 (2003)
13. Justi, R.S., Gilbert, J.K.: Science teachers’ knowledge about and attitudes towards the use of
models and modelling in learning science. Int. J. Sci. Educ. 24(12), 1273–1292 (2002)
14. Lee, E., Brown, M.N., Luft, J.A., Roehrig, G.H.: Assessing beginning secondary science
teachers’ PCK: pilot year results. School Sci. Math. 107(2), 52–60 (2007)
15. Magnusson, S., Krajcik, J., Borko, H.: Nature, sources, and development of pedagogical
content knowledge for science teaching. In: Gess-Newsome, J., Lederman, N.G. (eds.)
Examining Pedagogical Content Knowledge, pp. 95–132. Kluwer (1999)
16. Rahimi, E., Barendsen, E., Henze, I.: Typifying informatics teachers’ PCK of designing
digital artefacts in dutch upper secondary education. In: Brodnik, A., Tort, F. (eds.) ISSEP
2016. LNCS, vol. 9973, pp. 65–77. Springer, Cham (2016). https://doi.org/10.1007/978-3-
319-46747-4_6
17. Saeli, M.: Teaching programming for secondary school: a pedagogical content knowledge
base approach. Ph.D., Eindhoven Institute of Technology (2012)
18. Sanders, L.R., Borko, H., David Lockard, J.: Secondary science teachers’ knowledge base
when teaching science courses in and out of their area of certification. J. Res. Sci. Teach.
30(7), 723–736 (1993)
A Notation for Sets, Sequences and Series
Possible Benefits for Understanding and Use
Susanna H. du Plessis and Vreda Pieterse(B)
University of Pretoria, Pretoria, South Africa

vreda.pieterse@up.ac.za
Abstract. The use of a well-designed and uniform notational system is

imperative to improve communication and understanding. Currently, a
fairly consistent notational system is used to notate the elements of sets,
but the notations for sequences, series and other quantifications often
vary from one author to another. The most widely used notation for sets
has limitations and cannot easily be generalised to sequences and series.
Furthermore, the well-known sigma notation for series is ambiguous and
difficult to grasp. This article proposes a clear notation that can be used
for all of the above. We aim to test the hypothesis that this notation
could aid an understanding of concepts which build on the use of this
notation. As a first step, we conducted a pilot study to inform the design
of an experiment to evaluate the effect of the proposed notation on under-
standing. Secondary-school learners who had never been exposed to the
sigma notation participated in the pilot study.
1 Introduction
Mathematical notation is a language in its own right – unique and constantly
evolving. Mathematicians, speaking a range of indigenous languages, use it to
express mathematical concepts in a way which everyone can understand. The
use of mathematical language has evolved over many centuries as users of this
language share a mutual understanding of the meaning of its symbols, words
and sentences. Good notation enhances the precision of expression and at the
same time simplifies communication.
Mathematical notations are introduced to provide concise and accurate ways
of communicating complex yet well-understood concepts. It is well known that
brevity is the leading characteristic of mathematical elegance, but this is not
the only requirement. Most mathematicians agree on the value of clever nota-
tions. Dijkstra and Van Gasteren [1] emphasise that the use of appropriate nota-
tion can make a difference in mathematical work. Lipton [2] gives an example
of the European mathematicians who used Leibniz’s dx dt differential notation,
which enabled them to progress faster than their British counterparts who used
Newton’s ẋ to express the same concept. To support mathematical think-
ing, notation should not only be designed to enhance the brevity of the text
but should also control the number of rules governing the manipulation of
expressions [1].
https://doi.org/10.1007/978-3-319-71483-7_7
78 S.H. du Plessis and V. Pieterse
The influence of notational differences on the ease of mastering the underlying

concepts, has been investigated in various contexts [3–5]. Chirume [6] found in a
study involving secondary schools in Zimbabwe, that a clear notation plays a role
at the initial stage of learning a new concept. He found that students might fail
to grasp mathematical concepts because they take the symbols themselves as the
objects of mathematics rather than the ideas and processes which they represent.
These studies confirm that clear and unambiguous notation may promote better
cognition and that issues of symbol familiarity and symbol density should be
considered when teaching mathematics [7].
Electronic recognition of mathematical expression is a large area of research,
one of which is the CROHME project that strives to build a comprehensive
database of handwritten mathematical expressions which can be used for the
online recognition of mathematical expressions. Other examples of such research
are Simistira et al. [8] that propose an algorithm using probabilistic SVMs and
stochastic context-free grammars to arrive at a marginal improvement in the
recognition of mathematical expressions in order to produce MathML output.
More specific research in the use of mathematical expressions online is that of
Cuartero-Olivera et al. [9]. They found that a crucial factor in the use of online
tools for mathematical expression is the time it takes to enter the expression.
They suggest a speech-to-text tool to circumvent this issue.
One of the notations that has been identified as particularly cumbersome,
when encountered for the first time, is Euler’s sigma notation [10]. This nota-
tion is usually introduced for the first time to South African learners at the
beginning of Grade 12 when covering the topic of sequences and series. This
paper proposes a consistent notational system for sets, sequences and series,
which has the potential to increase familiarity and reduce the introduction of
additional symbols when teaching these topics. We aim to investigate the value
of using the proposed notations: firstly to simplify the electronic formulation of
the constructs and secondly, to promote an understanding of the concepts. The
participants in the pilot study were secondary-school Grade 11 learners who had
not previously been introduced to series or to the use of the sigma notation.
2 Sets
A set is a collection of objects. One way of describing or specifying the members

of a set is by extension, i.e. by listing each member of the set. When the members
of a set follow a pattern, the set can be specified by intentional definition, i.e.
by using a rule or semantic description as specified in the international standard
ISO 80000-2:2009 [11]. The general format for the notation is:
{x | P (x)}
The symbol | means such that. Therefore, the above means the set of all elements
x such that P (x) is a true statement. The property P can be expressed in either
words or symbols. The variable on the left of the | specifies a dummy whereas
A Notation for Sets, Sequences and Series 79
the expression on the right delineates the scope. One may use formulas to specify
the dummy. For example, the set of odd natural numbers can be expressed as:
{2t + 1 | t ∈ N}
Dijkstra [12] criticises this notation. He gives the following example which reveals
an inherent flaw in this notation:
{in | i < n}
It can be interpreted as {1n , 2n , 3n , . . . , nn } or as {ii+1 , ii+2 , ii+3 , . . . }. This

ambiguity arises because the specification of the dummy is not separated from
the description of the elements. He proposes a notational system to remedy this
deficiency. This notation requires the clear separation of three aspects, namely
(i) the dummy elements, (ii) the scope description, and (iii) the description of the
elements of the set in terms of the dummy elements. Dijkstra’s notation for the
intentional definition of a set specifies these aspects separated by the: character
and enclosed in angle brackets. This notation not only removes ambiguities, it is
also a more versatile expression of the conditions for membership of a set. The
following is the general format for Dijkstra’s notation:
x : P (x) : f (x)
Here x is the dummy, P (x) is a predicate that specifies the scope and f (x) is
an expression that describes the elements of the set. More than one dummy, as
well as more than one predicate to specify the scope, may be used. When doing
so, they should be separated by commas. This allows a distinction between the
following:
i : i < n : in , n : i < n : in and even i, n : i < n : in
Dijkstra states that he has no logical objection to declaring the type of the
dummy when identifying the dummy. For example, the following are equivalent
specifications of the set of even numbers less than 100 using Dijkstra’s notation:
i ∈ N : i < 100 : 2 × i
i : i ∈ N, i < 100 : 2 × i
In this paper, we introduce a notation similar to Dijkstra’s notation but

closer to the ISO standard. The following is the general format for the proposed
notation:
{ x | P (x) | f (x) }
An obvious difference between this notation and Dijkstra’s is the use of the
punctuation prescribed in the ISO standard, i.e. { | | } instead of : : . To
avoid further ambiguities the following restrictions are also proposed:
– Type specifications of dummy variables are not allowed.
– Multiple scope predicates are not allowed.
If needed, the type of a dummy variable may be specified by including it in the

scope predicate. Limiting the specification of the scope to a single predicate is
not a real restriction, because if both P1 and P2 should hold, it can just as well be
specified with the single predicate P1 ∧ P2 where ∧ is the symbol for the Boolean
AND operation. The following is therefore the only legitimate specification of
the set of even numbers less than 100, using the proposed notation:
{ i | (i ∈ N) ∧ (i < 100) | 2 × i }
3 Sequences
A sequence is an ordered list of objects. A sequence differs from a set because the
order of the objects matters. In addition, exactly the same elements can appear
multiple times at different positions in the sequence. A sequence with n entries
is called an n-tuple. As far as we know, no standard has been specified to notate
an intentional definition of sequences. In this paper we propose a notation for
cases where there is a relationship between i and the value of the ith term in
the sequence. This is an adaptation of the proposed notation for the intentional
definition of sets in Sect. 2. Similar to the notation for sets, the three aspects —
namely the dummy variables, the range predicate and the expression describing
the entries — are specified and separated by the | character. This is a variation of
the notation proposed by Pieterse [13]. Here we use the punctuation prescribed
in the ISO standard for sequences.
To indicate that it is a series and not a set, parentheses are used instead of
curly braces, for example, the following specifies the sextuple (3, 5, 7, 9, 11, 13).
The formula for the ith term in this sextuple is 2 × i + 3. It can therefore be
described by using the following intentional definition:
( i | i ∈ N6 | 2 × i + 3 )
4 Series
A series is the sum of the terms of a sequence. The well-known sigma notation
for the summation of sequences was introduced by Leonard Euler in 1755 [14].
Euler’s notation and modern refinements of this notation are well established in
the mathematical community. When using Euler’s notation, 32 +42 +52 +62 +72
is written as:
7
m2
m=3
Wees [10] contends that students find Euler’s sigma notation difficult to
understand at first. He attributes this to the complexity of the expression,
which is a function that takes as many as four parameters, all of which have
to be understood at once. He proposes a programming-like notation as a substi-
tute for Euler’s sigma notation. His notation requires descriptive names for the
parameters. Though his proposal contributes significantly to the clarity of the

expression, it loses two essential attributes of viable notational systems, namely
conciseness and independence from natural language.
Dijkstra [12] points out a flaw in Euler’s notation which is beyond the prob-
lem observed by Wees, namely an inherent semantic ambiguity resulting from
uncertainty related to the extent of the description of the elements in the series,
for example,

7
7
7
m2 + 1 can be interpreted as m2 +1 or as (m2 + 1)
m=3 m=3 m=3
This uncertainty can be resolved by introducing parentheses. Dijkstra, how-

ever, proposes that the use of an adaptation of his set notation should replace
Euler’s notation. Dijkstra’s proposed notation enforces careful thought which
would eliminate possible ambiguities, whereas Euler’s notation allows ambigu-
ous expression, placing the onus on the writer to solve possible ambiguities.
Dijkstra follows Euler’s idea of using the Σ symbol to indicate summation. In
Dijkstra’s proposal, this symbol is specified along with the dummy variable.
The above-mentioned ambiguous specification of a series can therefore only be
specified as one of the following unambiguous expressions:
Σm : 3 ≤ m ≤ 7 : m2 + 1 or as Σm : 3 ≤ m ≤ 7 : m2 + 1
Besides promoting unambiguity of expression, Dijkstra’s notation reduces

the need for spatial and size differences to convey meaning in comparison with
Euler’s notation. Dijkstra’s notation has the advantage of clearly distinguishing
the description of the terms in the sequence from the surrounding text. The
linearity of the Dijkstra notation will make electronic recognition simpler and
faster. It does, however, have drawbacks, namely the introduction of an unfamil-
iar symbol, overloading the purpose of a section in the expression, and applying
the operation in the wrong context. We discuss these problems and state how
we avoid each of these drawbacks when proposing our own notation.
4.1 Introduction of an Unfamiliar Symbol
Both Euler and Dijkstra use the Σ symbol to indicate summation. This is an
entirely new and unfamiliar character for learners who have not been introduced
to this topic. We avoid the introduction a new symbol by simply using the +
symbol. The learners are already familiar with the symbol and its meaning.
4.2 Overloading the Purpose of a Section in the Expression
Dijkstra overloads the purpose of the first section of the expression when requir-
ing the specification of an operation along with the dummy variable in this
section. We suggest that the symbol that specifies the operation should be placed
as a prefix operation instead of embedding it in the specification.
4.3 Applying the Operation in the Wrong Context

Dijkstra amended the expression for a set to serve as an expression of a series.
Interpreting the series as the summation of a set poses some problems. Consider
the following series:

10
(i % 2)
i=1
where % is the symbol for the modulus operation, namely to determine the
remainder when one divides the value of the left expression by the value of the
right expression. The value of the this series is 5. This is determined by evaluating
the following expression:
1+0+1+0+1+0+1+0+1+0
When representing this series using Dijkstra’s notation, one would write
Σ i : 1 ≤ i ≤ 10 : i % 2
When interpreting the expression – assuming that the summation is over the
set which may only have unique values – the expression is equivalent to the
summation of the elements of {0, 1}. The resulting value is 1 instead of the
intended 5. This error in interpretation arises from the assumption that the
specified terms are elements of a set, instead of the intended meaning, namely
to be the terms in a sequence, in which the same element may appear multiple
times. To remedy this problem, our proposed notation amends the expression
for a sequence to serve as an expression of a series.
4.4 Proposed Notation

Our notation is similar to Dijkstra’s notation, yet it has the following differences:
– It uses the + symbol instead of the Σ symbol to denote summation.
– The operation symbol is written as a prefix to the sequence instead of placing
it along with the specification of the dummy variable.
– It uses parentheses where Dijkstra’s notation uses angle brackets. This nota-
tion therefore applies explicitly to series (not sets) and uses the punctuation
prescribed for sequences in the ISO standard.
The following is the general format of the proposed notation for series:
+( x | P (x) | f (x) )
The meaning of the parameters is the same as in the notation for sequences: x is
the dummy, P (x) is a predicate that specifies the scope and f (x) is an expression
that describes the terms in the series. For example, when using our notation,
the expression

7
(m2 + 1) is written as + ( m | 3 ≤ m ≤ 7 | m2 + 1 ).
m=3
The proposed notation is more versatile than other notations. It does not intro-
duce a new symbol and can therefore be used without adaptation for quantifica-
tions involving operations other than +. Other notations require the introduction
of additional symbols when used for quantifications involving other operations.
For example, when using Euler’s notation, the symbol Π is introduced to indicate
multiplication over a series using, but our notation simply uses ×.
Owing to its linear nature and the absence of non-standard keyboard char-
acters the proposed notation may pose fewer problems when used electronically.
5 Pilot Study
We designed a pilot study [15]. The purpose of the pilot study was to gain
insight into the use of the notation in the field and to evaluate the feasibility
of conducting a full-scale investigation to determine the effect of the use of
our proposed notation on the participants’ comprehension of the mathematical
concepts underlying sequences and series.
The participants in the study were Grade 11 learners who had never been
exposed to the topics in question. It is unlikely that these learners would have
been exposed to the sigma notation elsewhere, since the target school follows a
rigid syllabus and the first introduction of the topic is in Grade 12. We divided
the learners randomly in two groups.
We used the same basic material from the Grade 12 mathematics textbook
published by Siyavula1 for both groups. The one group’s material was modified
to use the Dijkstra notation but used curly brackets instead of angle brackets.
The same teacher presented the material to both groups to avoid differences in
presentation. We did not introduce more notations because we are not aware of
any other notations for these constructs and we did not have enough participants
to conduct a meaningful experiment involving more notations.
The experiment was conducted over a period of three days with a work period
of approximately three hours every day. The teacher was an experienced lecturer
who had taught mathematics in secondary school for more than five years, was
involved in teacher training and was familiar with both the notations. The same
teaching aids were used in both classes. The lessons involved verbal explanations
combined with questions to explain the concepts. Each lesson was concluded with
use of worksheets containing exercises which the learners did on an individual
basis. The teacher moved from one classroom to the next, presenting the same
topic but using a different notation. While one group was receiving a lesson, the
other group was doing exercises.
After all the material had been covered, both groups wrote the same test
(Test 1). We then inverted the groups, giving the first group a view of the Dijkstra
notation and the second group a view of the traditional (Sigma) notation, and
again gave the same test to both groups (Test 2), requesting that they use the
notation of their choice to write their answers. The final test was intended to
ensure that the learners were able to use both notations.
1
https://www.siyavula.com/maths/grade-12.
At the end of the three days, an opinion poll was held, involving the par-
ticipants who had remained in the experiment until the end. We asked four
questions in this opinion poll:
1. Which notation is easier to write?
2. Which notation is easier to read?
3. Which notation is easier to understand?
4. Which notation do you prefer?
5.1 Learners’ Performance

Table 1 shows the descriptive statistics of the marks that the learners achieved
in the tests, while the individual performance of each participant is shown in
detail in Figs. 1 and 2.
Table 1. Descriptive statistics
Test 1 Test 2
Average Median Std Dev Average Median Std Dev
Sigma group 37.00% 37.70% 0.09 20.71% 19.19% 0.09
Dijkstra group 17.21% 13.93% 0.17 13.30% 7.07% 0.11
Test 1
60 Test 2
40
Percentage
20
D1 D2 D3 D4 D5 D6 D7 D8 D9 D10 D11 D12

Participants
Fig. 1. Dijkstra group test results
The results of the final test were lower than the first, which is unexpected,
since one would imagine that learners would perform better after having had
60 Test 1
Test 2
40
Percentage
20
S1 S2 S3 S4 S5 S6 S7 S8 S9 S10 S11 S12 S13 S14

Participants
Fig. 2. Sigma group test results
more exposure to the underlying concepts. We can only theorise that the lack of
performance was due to the learners having had little incentive to complete the
tests, so the learners did not put much effort into the final test.
Statistical analysis is not feasible in this case, due to the small number of
participants and the large deviation in the results. The one group was clearly
stronger than the other. Evidently our selection method was flawed as it did not
render comparable groups.
A factor which might have played a role is that the two groups were not
equal in gender distribution. To make matters worse, the dropout rate over
the three days skewed the gender distribution even more (Table 2). The group
with a majority of girls performed much better on average than the group with
predominantly boys. The difference in the performance of the groups is so obvious
that it is unlikely to be coincidental. The reason for this difference is, however,
unclear. We are inclined to attribute it to gender differences rather than the way
in which the material was presented in this experiment. This opinion is a topic
for another investigation, which is beyond the scope of the present research.
Table 2. Start and finish number of participants
Start Finish
Girls Boys Total Girls Boys Total
Sigma group 11 5 16 10 4 14
Dijkstra group 5 12 17 4 8 12
Total 16 17 33 14 12 26
Table 3. Opinion poll results
Sigma Dijkstra No Sigma Dijkstra No

notation notation preference notation notation preference
Easier to write Easier to read
Sigma group 8 5 1 10 4 0
Total 18 6 2 17 8 1
Easier to understand Preferred
Sigma group 9 4 1 10 4 0
Total 17 7 2 18 8 0
5.2 Opinion Poll Results

The results of the opinion poll were inconclusive but quite interesting, since the
stronger academic group had a higher number of learners who preferred the
new notation, although they had been primarily instructed in using the Sigma
notation. This indicates that there might be merit in conducting a full-scale
experiment to obtain more reliable results.
Table 3 shows the notational preferences of the 26 participants who com-
pleted the experiment. The majority of the learners favoured the sigma notation.
Regardless of which notation they were taught at first, 18 preferred the Sigma
notation and 8 the Dijkstra notation.
The following can be observed:
– 35% of the group instructed by using the Sigma notation felt that the Dijkstra
notation was easier to write.
– 66% of the group instructed by using the Dijkstra notation preferred the
Sigma notation.
All the tests and exercises were done in a hand-written format. It is clear that
the participating learners preferred the sigma notation when they had to use pen
and paper. This could be due to the easier visual separation of the elements in
this notation but more research would have to be conducted before we could
draw conclusions about this preference.
5.3 General Comments
The logistical issues involved in organising a study of this kind are daunting. Per-
mission has to be obtained from governing bodies and teachers, and the parents
also have to give their consent. The permissions alone could take several months.
Since the school where we conducted our research, is semi-private and depen-
dent on parental funding and goodwill, it is understandable that the schools was
reluctant to introduce anything that could be seen as remotely controversial.
Once all the stakeholders had been persuaded and the requisite permissions
obtained, a suitable time slot had to be found in the school’s extremely packed
agenda. For our experiment, this date ended up being after the final exams
just before the summer holidays. In the South African school system, learners
who have completed their end-of-year examinations do not wait for the official
closing date of the schools to go on holiday. This meant that our pool of available
learners was small and that the participating learners’ motivation to sit through
lessons and tests was very low. This could also explain the high dropout rate.
The participating learners had on average a scanty understanding of the
material presented and none of the results could be used to draw conclusions
about the influence that the use of the different notations had on the ease of
comprehension of the underlying concepts.
6 Future Research
In the process of analysing the results of this experiment, several observations
were made that warrant further research into the potential benefits of a new
notation for the sum of sequences.
We have decided to conduct our full-scale research experiment online. This
will enable us to reach a wider audience and to track learner progress more easily.
Taking the research online will also allow us to improve some of the aspects of
classroom teaching as experienced during this research.
7 Conclusion
Inventing a useful notational system that is economical and aesthetic is an art.

A notational system should be equally convenient to write by hand and to typeset
with the use of contemporary tools. If it is not promoted in the right place at the
right time, it may remain unnoticed and unused. Abadir [16] concedes that it is
likely that authors will not adhere to proposed notational standards. He uses the
example of Bernoulli [17] who did not adopt the = sign for equality 150 years after
it had been proposed, even though many other mathematicians used it.
In this paper, we propose an alternative notation for series to replace the
widely used yet cumbersome sigma notation. Our notation is an amendment of
a notation designed by Dijkstra [12]. He proposed a set notation to avoid some
flaws that are inherent in the standard set notation. He re-appropriated his set
notation to specify series. We discuss problems with Dijkstra’s notations and
propose our own notations for sets, sequences and series, which we believe are
more elegant, user friendly and robust than other known notations for these
constructs. The notation might enhance understanding of the underlying math-
ematical concepts.
The results of our pilot study were inconclusive. There could be several rea-
sons for this but we believe it was mainly because we had too little time to convey
the topics sufficiently well for the learners to grasp the subject meaningfully.
We are still hopeful that additional research will show that the new notation
is beneficial to learning and also more practical in electronic use and e-learning
programs.
References
1. Dijkstra, E.W., van Gasteren, A.J.M.: On notation, January 1986. http://www.cs.
utexas.edu/users/EWD/ewd09xx/EWD950a.PDF. Accessed 27 Nov 2013
2. Lipton, R.: Notation and thinking, November 2010. http://rjlipton.wordpress.com/
2010/11/30/notation-and-thinking/. Accessed 12 May 2016
3. Hoch, M., Dreyfus, T.: Structure sense in high school algebra: the effect of brack-
ets. In: Proceedings of the 28th Conference of the International Group for the
Psychology of Mathematics Education, Bergen University College (2004)
4. Kisiel, V.M.: I saw the sign and it opened up my eyes I saw the sign! A study
of the impacts of the use of different multiplication symbols in the mathemat-
ics classroom. Master’s thesis, Department of Mathematical Sciences at the State
University of New York at Fredonia, Fredonia, New York, July 2010
5. Torigoe, E., Gladding, G.: Symbols: weapons of math destruction. AIP Conf. Proc.
951(1), 200–203 (2007)
6. Chirume, S.: How does the use of mathematical symbols influence understanding
of mathematical concepts by secondary school students. Int. J. Soc. Sci. Educ. 3(1)
(2012). ISSN: 2223–4934 E and 2227-393X
7. Bardini, C., Pierce, R.: Assumed mathematics knowledge: the challenge of sym-
bols. Int. J. Innov. Sci. Math. Educ. 23(1), 1–9 (2015). Formerly CAL-laborate
International
8. Simistira, F., Katsourosa, V., Carayannis, G.: Recognition of online handwritten
mathematical formulas using probabilistic SVMs and stochastic context free gram-
mars. Pattern Recogn. Lett. 53, 85–92 (2014)
9. Cuartero-Olivera, J., Hunter, G., Prez-Navarro, A.: Reading and writing mathe-
matical notation in e-learning environments. ELC Research Paper Series Issue 4
(2012)
10. Wees, D.: Mathematical notation is broken, May 2012. http://davidwees.com/
content/mathematical-notation-broken. Accessed 24 Nov 2013
11. ISO/TC 12: ISO 80000–2:2009 Quantities and units – Part 2: Mathematical signs
and symbols to be used in the natural sciences and technology, November 2009.
http://www.iso.org/iso/catalogue detail.htm?csnumber=31887. Accessed 22 Dec
2013
12. Dijkstra, E.W.: EWD1300: The notational conventions I adopted, and why. Formal
Aspects Comput. 14, 99–107 (2002)
13. Pieterse, V.: Topic maps for specifying algorithm taxonomies: a case study using
transitive closure algorithms. Ph.D. thesis, University of Pretoria (2017)
14. Euler, L.: Foundations of differential calculus. Transl. by Blanton, J.D. Springer,
New York (1755/2000)
15. du Plessis, S.H., Pieterse, V.: Die effek vann alternatiewe notasie op die begrip van
wiskundige konsepte vir graad 12 leerders. Suid-Afrikaanse Tydskrif vir Natuur-
wetenskap en Tegnologie 36(1), 1–2 (2017)
16. Abadir, K., Magnus, J.: Notation in econometrics: a proposal for a standard. Econo-
metrics J. 5(1), 76–90 (2002)
17. Bernoulli, J.: Ars conjectandi. Impensis Thurnisiorum, fratrum (1713)
Bebras Tasks
Bebras Task Analysis in Category Little Beavers
in Slovakia
Lucia Budinská(B) , Karolı́na Mayerová, and Michaela Veselovská
Department of Informatics Education, Faculty of Mathematics,

Physics and Informatics, Comenius University in Bratislava, Bratislava, Slovakia
{lucia.budinska,karolina.mayerova,michaela.veselovska}@fmph.uniba.sk
Abstract. In Slovakia, there has been an international Bebras compe-

tition (iBobor) since 2007. The first category named Little Beavers (Pri-
mary or Bobrı́ci) includes pupils aged between 7 and 10 (2nd - 4th grade
of primary school). In this article, we analyse tasks and results from this
category collected in years 2012–2017. By qualitative and quantitative
research methods, we have created a new categorisation of tasks based
on their text analyses in order to better understand some patterns (and
correlations) between the tasks and the results. We have created four dif-
ferent categories: Programming oriented tasks, Algorithmic, Logic and
Digital Literacy. Using qualitative task analysis, we have also defined
several subcategories and, based on analysis of contestants’ results, we
have found different grade- and gender-to-performance correlation.
Keywords: Bebras competition · Primary education · Categorisation

task analysis · Gender performance
1 Introduction
In the school year 2016/2017 we participated in the 10th year of the competition
iBobor1 in Slovakia. It is an international competition named Bebras which
originated in Lihuania and it is focused on informatic tasks [1]. In Slovakia iBobor
competition has been organized since the school year 2007/2008. We included
the category named Little Beaver (at the official website of competition2 named
Primary) into competition iBobor in the school year 2011/2012 for the first time
[2]. It was the first year when the fourth grade pupils had already had Elementary
Informatics during primary school. This compulsory school subject has been
taught since the school year 2009/2010 from the second to the fourth grade for
one school period (45 min) per week. The new educational reform introduced
in 2014 renamed aforementioned subject to Informatics [3] and placed it in the
third and the fourth grade at primary school. In the first year, there were 7,727
competitors in category Little Beavers. Since then, the number of participants
has gradually increased, and this year it was 15,486. The category Little Beavers
1
see http://ibobor.sk/ - in Slovak only.
2
see http://www.bebras.org/.
https://doi.org/10.1007/978-3-319-71483-7_8
92 L. Budinská et al.
is intended for the third and the fourth grade pupils at primary schools. However,
teachers may also register younger pupils into this category. Due to the age and
cognitive development of the primary school pupils, the organizers have decided
to reduce the number of tasks to 12 for this category along with the reduction of
time to solve the tasks down to 30 min, while other categories include 15 tasks
each with 40 min to find the solution. Tasks in this category typically contain
less text information and more pictures. The tasks mentioned are in the form
of stories or they focus on real-life situations that should be well known to
pupils - so it tries to eliminate abstraction [2].
2 Research Problem
Over the past few years, we have gained a great deal of data that can help us to
create a picture of the state of pupils’ knowledge. In this article, we focus on the
Little Beavers category and the results of pupils in this category over the last
5 years. Tasks in this competition usually come from an international database,
yet about 50% of the used tasks in this category have been created by Slovak
authors (Fig. 1). The reason may be that not all countries have a category for
the primary education in the competition. Within the international database [4],
the tasks are assigned to one or more of the following areas:
– Information comprehension,
– Algorithmic thinking,
– Using computer systems,
– Structures, patterns and arrangements,
– Puzzles,
– Social, ethical, cultural, international, and legal issues.
Fig. 1. Authorship of tasks in last five year of iBobor competition in category Little
Beavers
Bebras Task Analysis in Category Little Beavers in Slovakia 93
However, the tasks used in Slovakia in the category Little Beavers are
designed according to the State Educational Program (hereinafter referred to
as SEP) [3], which is divided into five thematic areas:
– representations and tools (work with graphics, work with text, work with
stories, multimedia work, information, structures);
– communication and collaboration (working with a web site, searching the
web, working with communication tools);
– algorithmic problem solving (problem analysis, interactive solution compila-
tion, solution by command sequence, interpretation of solution writing, find-
ing and correcting errors);
– software and hardware (work with files and folders, work in the operating
system, computer and add-ons, work on a computer network and on the
Internet),
– information society (safety and risks, digital technologies in the company,
legality of use).
The tasks categories used in the international database overlap only to a cer-
tain extent with the tasks originating from thematic areas of the SEP. Therefore,
we are not able to uniquely include the tasks created for competition in the con-
text of informatics in Slovakia. In 2009 Kalaš and Tomcsányiová proposed new
categorisation for Bebras task [5] consisting of four categories: digital literacy,
programing, problem solving and data handling, but categories are broad and
overlapping as one task can fit into one or two categories.
In [6] the new two-dimensional categorisation system for tasks in Bebras
contest has been introduced. It combines computational thinking skills with
informatics concepts. Each task can be placed in only one informatics area, but
in up to three computational thinking areas.
One of the objectives of this article was to investigate the success of pupils
depending on different phenomena (e.g., conceptual complexity of the task, type
of the task, age and gender of the contestant, etc.) and we therefore created
our own categorisation to examine these dependencies. The base of it is similar
to [5], but we tried to make unambiguous categories and define subcategories
whenever possible.
There are not so many researches aimed at Little Beavers category, as many
countries don’t have this category, but overal we can say, that in most countries
boys and girls has similar performance in lower categories [5,7,9]. Boys tend to
be more successful in tasks aimed at spatial thinking, creating strategy and in
harder tasks, and girls perform better in tasks with colorful pictures and social
themes [8]. Contestants are more likely to guess the correct answer than to use
no respond answer, and boys do it more often (and even more successful) [7]. In
lower categories the proposed difficulty of tasks usually differ from real difficulty,
and pupils in this categories tend to underestimate the complexity of the tasks
[9]. Pupils in primary schools have problems with reading long texts, they cannot
focus on task for a long time, and need to have unambiguous texts and pictures
[2,9].
3 Methodology
The aim of our work was to better understand the results of contestants, and also
find out correlation between assignments of individual tasks and their results.
In our research, we have used qualitative and quantitative research methods.
When analyzing the task assignment, we used grounded theory with the system-
atic design [10]. We coded 60 tasks from the Little Beavers category over the last
5 years, and, based on them, we subsequently created new categorisation. We tried
to preserve the data triangulation and the objectivity of the results by individual
authors’ coding and subsequent joint processing based on inter-coder agreement
[11]. The created categories are described in Sect. 4. As in the competition, all tasks
are evaluated by points based on difficulty, using three groups: easy, medium and
hard, with 4 tasks being in each group. The proposed difficulty does not always
correspond to the pupils’ results, so for our research, we have redistributed the
analysed tasks into these three categories, depending on the real difficulty. (We
sorted tasks based on their results, first four with the best average score were
labeled as easy, next four as medium and four with the lowest score were hard.)
Next, we used qualitative methods to analyse data from the Slovak competi-
tion database, where the following information are kept about each contestant:
gender, grade, school’s ID, the chosen answer and the time they needed to solve
the contest. From this database, we first exported information about pupils com-
peting in the category Little Beavers, which we further processed using infer-
ential statistics for different groups - based on gender, or grade. We have also
used correlation designs [10], examining relations and dependencies between the
tasks results and the category in which they were placed.
4 New Categorisation
As it was mentioned above, we have created four new categories where all types
of tasks used in Slovak competition in category Little Beavers can be included.
When creating these categories, we focused on analysing the tasks’ text, or more
specifically their form and content, and the type of tasks (which area of knowl-
edge is tested). We have further analysed which information is available in text
and which is needed to be analysed by pupils, and also how they come up with
the answer - if they choose it or they need to create it. Example for each category
is shown in the Fig. 2.
4.1 Digital Literacy Tasks

Tasks oriented towards digital literacy are those that focus on verifying the skills
in using some (relatively general) software or hardware. We have divided them
into two subcategories:
i. basic - if the task is focused only on one type of skill or knowledge,
ii. combined- if the task contains some more difficult concept or combines basic
digital skills with algorithms, pattern recognition or rule identification.
Fig. 2. Examples of tasks for each category from iBobor in school year 2016/17 in
Slovakia. Volume - digital literacy task. Shelf Sort - logic task (statement). Bebras
painting - algorithmic task (rules are given, answer is created). Robot 1 - programming
oriented task (program creation).
This type of tasks are not usually used in Bebras in other countries and they
are not very supported by the international community. However, we have found
out that pupils improve between the third and the fourth grade, so we have
come to the conclusion that this area of informatics takes a big part in Slovak
education. In other countries like for example England, digital literacy is a part
of different subjects, not only informatics or computing.
4.2 Logic Tasks

By logic tasks we mean tasks where the steps or algorithm to be followed are
not clearly defined. The pupils use exclusion, matching or deduction to solve the
tasks. We divided them into three subcategories:
i. statement - Statement logic tasks are tasks in which pupils work with logic
statement usually given in text, sometimes in a graphical form. Using these
statements they deduce the answer.
ii. graphs - In the graph tasks, pupils work with the data structure repre-
sented by a graph. Typically, it is a tree, a graph (e.g. a square network) or
even a linked list or a diagram. More difficult tasks also include some graph
algorithm - for example minimal path finding.
iii. others - This is a minimal set of tasks that do not belong completely to either
of the two previous categories. It can be a mixture of the two, or it can be a
task demonstrating a completely different information representation. This
category also includes logic puzzles.
4.3 Algorithmic Tasks

These are the tasks in which a pupil needs to follow a procedure or an algorithm
or a set of instructions or rules to work with some objects or information to get
an answer. We have identified two additional subcategories for these tasks based
on the tasks’ rules and the way of response.
i. rules
a. are given - Pupils have to work with exactly defined conditions, rules, an
algorithm or steps to follow and it is obviously stated for them how to
solve the task.
b. need to be identified - Pupils are not explicitly told the algorithm or condi-
tions for solving the problem, so they need to identify them by themselves
in the text.
ii. answer
a. is created - In this categorisation we do not focus on the way the answer
is given in the competition (multi-choice or interactive), but on the way
how pupils come up with the answer. They can create the answer from a
scratch or simulate the whole algorithm, so that the answer is created.
b. is identified - Pupils are already given a pre-prepared answer and they
either need to finish it, or they need to identify the right answer which is
not needed to be created from a scratch.
4.4 Programming-Oriented Tasks

Tasks which clearly include some execution or a creation of a program or a
sequence of steps in a form of text, icons, pictures or combination of the previous
are included in this category. Although these tasks might be quite similar to
the algorithmic tasks, they specifically need to fall into one of the following
subcategories:
i. creation of a program - In this category pupils had to create or complete
a sequence of steps or a program.
ii. interpreting the program - In this category pupils need to interpret or
simulate a program or a sequence of steps written in text to find the correct
answer.
5 Results
First of all, we analysed data of the last five years of Little Beavers category for
grade and gender groups. The competition is more attractive to boys than to girls
and the average score of girls and boys is similar, with girls being slightly better
in four out of five years (see Fig. 3). Girls achieved better results in tasks where
it was not necessary to create or discover an algorithm or use some strategy, but
to perform some sequence of steps or commands or to evaluate some states and
statements.
Fig. 3. The last five years of iBobor competition in category Little Beavers - a. number
of contestants, b. overall performance for boys and girls (maximum score is 96 points,
minimum score is 0 points
There were more fourth graders than third graders in competition and on
average 2% of all pupils in category were second graders. Interestingly, the sec-
ond graders, who entered the contest despite the fact that they were in the cate-
gory with older pupils, achieved on average very similar results as third graders
(see Fig. 4) and they even outperformed third graders and sometimes even fourth
graders in some tasks. The reason may be that teachers only enrolled the most
clever second graders, but there were often entire classes enrolled in the compe-
tition in the third and fourth grades.
Fig. 4. The last five years of iBobor competition in category Little Beavers - a. number
of contestants in each grade, b. overall average performance for each grade (maximum
score is 96 points, minimum score is 0 points
Next, based on our new categorisation and data analysis, we were looking
for some correlations and dependencies. We took a look at the group of pupils
who contested in the school year 2015/16 as third-graders and again in 2016/17
as fourth-graders and compared them to the corresponding grade score average.

We found that our group improved mostly in digital literacy based tasks. There
could be many reasons why, but it is very likely that this confirms our belief that
informatics teachers mostly include digital literacy in their lessons and therefore
pupils are most advanced in developing user skills. The same pupils were slightly
better than other contestants in the logic and algorithmic task categories and in
the programming-oriented tasks they reached approximately the same level.
When comparing the results of all five years, we wanted to discover some cor-
relation that would help us to estimate the difficulty of the task before its appli-
cation in the competition. However, such correlation and dependencies could
be ambiguous, as Vanı́ček [12] declares, although at least some correlation had
been confirmed. These claims are divided into four groups according to the new
categorisation, see below. For each group we have created a scatter plot to see
if some gender correlation could be found (see Fig. 5).
Fig. 5. Correlation of gender performance for each group of tasks. Green line represents
the same score for boys and girls. Point below the line means that girls performed better
in representing task, point above the line means that boys were better. (Color figure
online)
5.1 Digital Literacy Tasks
– The differences between boys and girls in this group were minimal.
– If an algorithmic or logical problem is combined with testing user skills, tasks
are more difficult for pupils - they are not easy.
5.2 Logic Tasks
– If the task falls into the logic statement group and task text is in form text-
image with clearly defined rules, task is easy for pupils.
– This also applies to image-image tasks that have clear rules but they may not
contain another algorithm-complex problem.
– If the task has a small finite number of solutions, it turns out to be easy.
– If the task had more than one correct solution or contains more complicated
rules, it is medium difficult. They are mostly in image-image form.
– If pupils need to create answer while the task contains more complicated rules
consisting of a number of conditions that need to be identified, the task turns
out to be hard.
5.3 Algorithmic Tasks

– If rules are clearly defined in the task, it will be hard or medium difficult
for pupils. This may be caused by the fact that authors, seeing that the
task contains a more complex algorithmic problem, try not to make it more
complicated by identifying rules.
– On the other side, if the pupils had to identify the rules in the task, it usually
had a high success rate, i.e. about 70%. In these tasks, pupils usually did not
have to create a strategy to solve them.
5.4 Programming-Oriented Tasks
– If task contains a programming concept – such as a variable, a cycle, an

obstacle movement or conditions – and program commands are represented
by images or pupils need to interpret text commands, task are difficult.
6 Discussion and Conclusion
In this paper we have described new categorisation of the tasks for Little Beavers
group in iBobor (Bebras) contest. We have used this categorisation to analyse the
last five years of the competition in Slovakia. We have found some correlations
and dependencies, mentioned in Sect. 5, although we cannot definitely tell if some
causalities really exist. So in the next competition iBobor, in the school year
2017/18, we plan to use these dependencies to categorise tasks in the difficulty
groups, and, based on the pupils’ results, we would be able to better validate
our statements. Also, we would like to use our categorisation on tasks used in
other countries in similar age category to compare results of slovak pupils with
their peers in other countries.
Pupils’ good performance in logic tasks can be influenced by their math
lessons or even other subjects, so it would be helpful to do qualitative research
aimed at primary school teachers (which in Slovakia use to teach all subjects
in their class) and talk with them about tasks in Little Beaver category and
possible interdisciplinary relations. So we would find correlation not only with

informatics, but get a whole picture.
We were looking at the difficulty of the tasks using recategorisation as
follows - first four tasks with the best results were easy, next four medium and
the last four (with the lowest score) were hard. In further analysis we would like
to recategorise tasks’ difficulty based on the percentage results, so the categories
could contain more or less than 4 tasks, and then use this recategorisation to find
dependencies between our proposed tasks categories and the task’s difficulty.
In gender-based analysis, we have found out that girls and boys have over-
all the same score, but some differencies between the tasks exist. Girls usually
perform better in easier tasks, while boys are better in the hardest tasks. Yet
to find out in which categories girls or boys are better, we need to create more
specific categorisation and analyse the data.
In [12] interactive tasks showed up as easy, but in our data some of them
were hard, so in the following research we will try to extend our categorisation
to answer type used in the contest with subcategories for interactive tasks. All
the proposed improvements could help us better understand what causes the
difficulties in the contest and also allow us to create more suitable tasks for
Little Beavers category.
Acknowledgement. The research was supported by the project Granty Univerzity

Komenského (Comenius University grants) UK/372/2017.
References
1. Dagienė, V.: Information technology contests - introduction to computer science
in an attractive way. Inform. Educ. 5(1), 37–46 (2006)
2. Tomcsányiová, M., Tomcsányi, P.: Little beaver – a new bebras contest cate-
gory for children aged 8–9. In: Kalaš, I., Mittermeir, R.T. (eds.) ISSEP 2011.
LNCS, vol. 7013, pp. 201–212. Springer, Heidelberg (2011). https://doi.org/10.
1007/978-3-642-24722-4 18
3. Štátny pedagogický ústav: Informatika. Inovovaný Štátny vzdelávacı́ program
(2014). http://www.statpedu.sk/sites/default/files/dokumenty/inovovany-statny-
vzdelavaci-program/informatika nsv 2014.pdf. Accessed 29 May 2017
4. Dagienė, V., Futschek, G.: Bebras international contest on informatics and com-
puter literacy: criteria for good tasks. In: Mittermeir, R.T., Syslo, M.M. (eds.)
org/10.1007/978-3-540-69924-8 2
5. Kalaš, I., Tomcsányiová, M.: Students’ attitude to programming in modern infor-
matics. In: Proceedings 9th WCCE: World Conference on Computers in Education,
Paper-No 82 (2009)
6. Dagienė, V., Sentence, S., Stupuriene, G.: Developing a two-dimensional cate-
gorization system for educational tasks in informatics. Informatica 28(1), 23–44
(2017)
7. Dagiene, V., Mannila, L., Poranen, T., Rolandsson, L., Stupuriene, G.: Reasoning
on children’s cognitive skills in an informatics contest: findings and discoveries
from Finland, Lithuania, and Sweden. In: Gülbahar, Y., Karataş, E. (eds.) ISSEP
2014. LNCS, vol. 8730, pp. 66–77. Springer, Cham (2014). https://doi.org/10.1007/
978-3-319-09958-3 7
8. Hubwieser, P., Hubwieser, E., Graswald, D.: How to attract the girls: gender-
specific performance and motivation in the bebras challenge. In: Brodnik, A., Tort,
F. (eds.) ISSEP 2016. LNCS, vol. 9973, pp. 40–52. Springer, Cham (2016). https://
doi.org/10.1007/978-3-319-46747-4 4
9. Dagiene, V., Stupuriene, G.: Bebras - a sustainable community building model
for the concept based learning of informatics and computational thinking. Inform.
Educ. 15(1), 25–44 (2016)
10. Creswell, J.W.: Educational Research: Planning, Conducting, and Evaluating
Quantitative and Qualitative Research, 4th edn. Pearson Education, Boston
(2012). 650 p. ISBN 978-0-13-136739-5
11. Silverman, D.: Doing Qualitative Research: A Practical Handbook, 4th edn. SAGE,
Newcastle upon Tyne (2013). 633 p. ISBN 978-1-4462-6014-2
12. Vanı́ček, J.: What makes situational informatics tasks difficult? In: Brodnik, A.,
Tort, F. (eds.) ISSEP 2016. LNCS, vol. 9973, pp. 90–101. Springer, Cham (2016).
https://doi.org/10.1007/978-3-319-46747-4 8
Promoting Computational Thinking Skills:
Would You Use this Bebras Task?
Annalisa Calcagni, Violetta Lonati, Dario Malchiodi, Mattia Monga(B) ,

and Anna Morpurgo
Università degli Studi di Milano, Milan, Italy

mattia.monga@unimi.it
http://aladdin.di.unimi.it
Abstract. Bebras, an international challenge organized on an annual

basis in several countries (50 in 2016), has the goal of promoting infor-
matics and computational thinking through attractive tasks. We ana-
lyzed Bebras tasks by considering the Computational Thinking (CT)
skills each task promotes, starting from the operational definition of
CT developed by the International Society for Technology in Education
(ISTE) and the ACM-founded Computer Science Teachers Association
(CSTA). We argue that such an approach is indeed well-suited to present
Bebras tasks, especially with the goal to use them in curricular teaching:
framing them as CT enablers helps in making explicit their educational
potential, that can be appreciated also by teachers without a formal edu-
cation in informatics and adapted to a wide range of ages and schools.
We explored the viability of our approach by interviewing teachers of
different kinds of schools. We propose to use these CT skills also as a
tool to classify Bebras tasks, which results in a more uniform distribution
of tasks, w.r.t. the one obtained by leveraging content topics.
Keywords: Informatics and education · Computational thinking ·

Learning contests · Bebras
1 Introduction
The last decade has seen an increasing need for spreading the fundamental con-
cepts of informatics to a vast audience of students, stemming from the belief that
some basic concepts of the discipline should be taught even in the first stages
of the educational systems. An important contribution to this goal is provided
by informatics contests organized worldwide: they are indeed able to stimulate
interest among pupils and teachers with different mixes of fun games and safe
levels of competitiveness [3]. An initiative that proved to be particularly success-
ful is the Bebras challenge1 [6,10,11], organized since 2004 on an annual basis
in several countries (50 in 2016), with about one and a half million participants
in the last edition.
1
http://bebras.org/.
https://doi.org/10.1007/978-3-319-71483-7_9
Promoting Computational Thinking Skills 103
The Bebras community gathers annually to discuss a pool of new tasks to

be proposed to pupils; from this pool the organizers in each country choose
the tasks to set up the local contests. The tasks should be fun and attractive,
be adequate for the contestants’ age and the solution should take on average
three minutes per task. Moreover, since the contest is aimed at a non-vocational
audience, tasks should be independent of specific curricular activities and avoid
the use of jargon. In fact, Bebras tasks focus on that part of informatics that
should be familiar to everyone, not just computing professionals. This computing
core is sometimes called computational thinking (CT) and its promotion is one
of the key goals in Bebras, whose full name is indeed “International Challenge
on Informatics and Computational Thinking” [8].
In the years since its inception, the Bebras community has developed hun-
dreds of tasks. Most of them are proposed as interactive and/or open-ended
questions. However, even when answers have to be chosen from a list, there is
no unique way of getting to the solution. The tasks can be used to organize new
contests, but they can also be the starting point for in-depth educational activ-
ities (a recent proposal is [8]). Potentially, their diversity represents a trove in
which every teacher could find suggestions and insights for introducing a compu-
tational topic or reflecting on it. Indeed, the Bebras community equips the tasks
with comments about their key points (“It’s informatics”). While some of the
Bebras countries (Switzerland, Lithuania, Singapore) provide additional mater-
ial to teachers, normally related to the latest edition (booklets or websites with
tasks, solutions, and an expanded version of their informatics context), in most
cases teachers just discuss the tasks with their students in one session after the
contest, and there is little evidence that teachers re-use older tasks and integrate
them in their curricular teaching activities (see, for instance [12]).
On the contrary, the Bebras corpus could become a considerable resource
to teach informatics and computational thinking, provided that tasks are made
easy to retrieve and their content is clearly signposted [8]. This statement is
consistent with a survey we conducted recently in our country (see Sect. 2).
After having considered other proposals for the classification of the Bebras
corpus (see Sect. 4), we decided to analyze it by means of the operational def-
inition of computational thinking [1] developed by ISTE (International Society
for Technology in Education) and CSTA (Computer Science Teachers Associa-
tion). We first identify the CT skills that are mentioned in the definition and
that are relevant when solving a Bebras task; for each of these skills we then
give a description that shows which kinds of tasks can promote such skill; finally
we analyse several tasks in order to detect the skills they promote. We argue
that such approach is indeed well-suited to present Bebras tasks since it helps
in making explicit their teaching potential. First, differently from classifications
based on a taxonomy of informatics content, this approach is more suitable for
identifying the cognitive skills involved in a task. Second, it can be appreciated
also by teachers without a formal education in informatics (still the majority,
in primary and non-vocational secondary schools): in fact the CT skill set uses
terms and concepts which can be grasped even without a deep knowledge of
104 A. Calcagni et al.
informatics technicalities. Finally, CT skills can be adapted to a wide range of

ages and schools, still maintaining their peculiarity, distinct from generic logi-
cal/analytical thinking.
We explored the viability of such approach by interviewing individually some
teachers of different kinds of schools. Teachers were provided with a short descrip-
tion [4] of the CT skills mentioned above and then were requested to associate
some Bebras tasks with those CT skills. More precisely, for each task and each
CT skill, we asked teachers the following question “Would you use this task to
promote such skill?”, and then discussed their answers with them, to under-
stand their motivations. The outcome of such interviews seems to confirm the
good potential of the approach (see Sect. 3).
We also propose to use this CT skills as a tool to classify Bebras tasks. We
manually classified the pool of tasks prepared for the 2016 contest (120 tasks)
and we verified that the resulting distribution among classes is more uniform
than that obtained by leveraging content topics.
The paper is organized as follows: in Sect. 2 we describe how the CT skills
mentioned in the ISTE/CSTA definition are related to Bebras tasks, and in
Sect. 3 we reports the findings of our interviews with teachers. In Sect. 4 we
apply this approach to obtain a classification of tasks according to the CT skills
they promote, and in Sect. 5 we draw some conclusions.
2 Bebras and CT Skills

After Bebras’s last edition (2016), we conducted a survey among the teachers
who had registered in the site of the contest in our country: an online question-
naire was filled in by 46% of those participating in the last edition (342 over
742 teachers); for most of them (71%) this was the first participation in Bebras
(w.r.t. the previous edition, the number of participant teachers has more than
doubled); in general we measured a strong appreciation for the initiative.
Among other questions, we were interested in understanding how teachers
make use of tasks (or intend to) in their classes. When asked whether they
intend to use Bebras tasks in their classes, 30.9% of the respondents answer
affirmatively (in particular 20.3% say that they’ve already used them); 65%
show interest in such a possibility but are uncertain (in particular 20.3% of the
respondents choose “I would like to, but I cannot figure out how”); negative
answers were marginal (3.6%).
A promising 43.7% of the respondents believe that Bebras tasks can be used
for curricular activities (the subjects more frequently mentioned are mathe-
matics, informatics, technology, depending on the grade and kind of school).
Moreover, we asked which options would make Bebras tasks easier to re-use for
teaching: besides some logistic issues concerning the access and form of online
and printed materials, the most appreciated proposals are: to have examples of
teaching units that use Bebras tasks (69.2%); to have tasks categorized/arranged
according to the computational thinking skills they require and/or promote
(51.5%); to have tasks categorized/arranged according to the informatic theme
they refer to (46.4%).
Hence teachers are clearly eager to have descriptions of the tasks that could
help them in finding the ones more suitable to their goals: both CT and infor-
matics seem useful for them, but while all tasks traditionally have an “It’s infor-
matics” section designed to explain the informatic idea behind it, no effort is
currently made by Bebras authors to identify and highlight the cognitive skills
they require. Thus, we decided to focus on CT skills, which seem to us the core
educational value of Bebras tasks, since they keep their computational peculiar-
ity, while being accessible even without a deep knowledge of the more technical
details of the discipline of informatics.
According to the operational definition proposed by ISTE and CSTA [1], CT
is a problem-solving process that includes (but is not limited to) the following
characteristics.
(a) Formulating problems in a way that enables us to use a computer and other
tools to help solve them.
(b) Logically organizing and analyzing data.
(c) Representing data through abstractions such as models and simulations.
(d) Automating solutions through algorithmic thinking (a series of ordered
steps).
(e) Identifying, analyzing, and implementing possible solutions with the goal of
achieving the most efficient and effective combination of steps and resources.
(f) Generalizing and transferring this problem solving process to a wide variety
of problems.
The first and last skills in the definition (a and f) are almost never addressed
by a single Bebras task, mainly due to its brevity, even though the contest aims
at helping the development of such skills in the long-term, by working on and
solving several tasks. Instead, most Bebras tasks deal with organization, analysis,
representation of data (skills b and c), algorithmic thinking (skill d) or design,
analysis and implementation of algorithmic methods (skill e). Thus, we believe
such a definition can be fruitful for analyzing and describing tasks.
ISTE/CSTA also propose a vocabulary of CT skills with a progression chart
suggesting possible activities for different ages and subjects [2]. The intended
goal of the vocabulary is “to unpack the operational definition by listing CT con-
cepts implicit in the operational definition”. The vocabulary lists and explains
nine terms, giving example activities suitable for the age groups: Data Collec-
tion, Data Analysis, Data Representation, Problem Decomposition, Abstraction,
Algorithms & Procedures, Automation, Simulation, Parallelization. This level
of description, however, seems to enter in the explicit domain of the practice of
informatics: as such, it could provide further enrichment for the “It’s informat-
ics” section which accompanies each task, but it is less useful to highlight their
teaching potential. In general, different (and only partially overlapping) defini-
tions of CT exist (a good recent survey can be found in [5], which discusses also
frequent misconceptions of CT by primary teachers), but in this proposal we
decided to focus on the operational ISTE/CSTA’s definition as the one with the
right granularity to emphasize the specificity of computational cognitive skills
versus a generic analytical approach, while being accessible to teacher without

any formal education in informatics.
In the following we illustrate the CS skills mentioned in the ISTE/CSTA
operational definition by referring to Bebras tasks that can promote such skills.
Fig. 1. Four tasks suitable to teach how to logically analyze data (a), how to represent
information (b), how to identify strategies and analyse algorithmic solutions (c), and
how to implement algorithmic solutions (d).
2.1 Logically Organizing Data
Typical tasks that promote this skill deal with ways to organize data according
to given criteria, with (generally hidden) references to databases or set theory:
they may ask to execute a query over a table of records representing a set of
objects, to split a set of items into categories according to their characteristics,
or to pick the misplaced object in a figure.
Other tasks for this skill focus on organizing data so that they enjoy relevant
properties: that is the case for instance of cryptography, where we want a message
not to be understandable even if eavesdropped, data compression, where we
organize data in order make them easier to store or transmit, or correction codes,
where we add bits to detect or recover possible errors in the representation of
stored or transmitted data.
This skill is promoted also by tasks where data structures are used to organize
data for processing. For instance, a task may concern the use of binary search
trees to find quickly any element in a set, without considering the whole set;
similar examples can be done for heaps, queues, stacks, and so on. However, if a
data structure is used to represent an intrinsic property of the data (e.g., graphs
for binary relations, or trees for hierarchical relations), the task will be most
suitably related to the skill Representing information.
2.2 Logically Analyzing Data

Many tasks for this skill are perceived as “logical problems”, since they require
logical inference, deductive reasoning, and drawing conclusions about the data
presented in the task.
Other tasks ask to check whether the data of the problem satisfy certain
properties; often such properties are not straightforward but some reasoning,
accurate observations (e.g., recognizing patterns), or a systematic approach are
needed to come to the right conclusion. This kind of tasks are well represented
by “Walnut animals”, proposed by Czech delegates in 2015 and depicted in
Fig. 1(a): to solve the task pupils need to abstract from the specific features of
animals and consider only their structure, then pairs of isomorphic animals can
be matched by analyzing their properties like the number of nuts, or the number
of connections from each nut.
2.3 Representing Information

Typical tasks for this skill deal with the digital representation of numbers,
images, colors, and sets of objects in general, or the visual representation of
data with diagrams like histograms or charts.
Other tasks introduce data structures to represent relevant properties of the
data, e.g., graphs for binary relations, or trees for hierarchical relations. For
instance, consider “Popularity”, proposed by Canadian delegates in 2015 and
depicted in Fig. 1(b) where a graph is used to represent and visually show the
friendship relation of people in a social network.
2.4 Algorithmic Thinking

Algorithmic thinking deals with transforming an intuitive idea into a form suited
to be processed automatically. Hence it allows, for instance: to design a system-
atic method to tackle a problem, starting from an intuitive approach; to trans-
form an intuitive idea of how to accomplish a task into a step-by-step procedure
that achieves the goal; to give a synthetic description of a situation or a process,
by detecting and exploiting its patterns and regularities; to start from a set of
examples or an informal description and formalize a rule that can be applied in
general; and so on.
Thus, typical tasks that promote this skill require to manipulate data follow-
ing a formal procedure (i.e., a sequence of ordered steps or moves) or a set of
rules or primitives. Tasks may require to execute some procedure or to compute

or recognize its output; to apply some transition rules to a system in a given con-
figuration; to predict the final state of a process described by a diagram (e.g.,
the transition diagram of a finite state automaton); to decompose a problem
into components; to combine primitive operations in order to compute a result
or accomplish a task; to systematically enumerate or examine all the possible
cases that can occur in a given context; and so on.
Notice that when the focus of the task is on implementing (analyzing, or
devising, respectively) an algorithmic solution, then the task should be most
suitably related to skill Implementing algorithmic solutions (Analyzing algorith-
mic solutions, or Identifying strategies, respectively).
2.5 Identifying Strategies
This skill concerns problem solving and in particular finding a suitable algorith-
mic strategy. Typically, formulating a solution algorithmically (so that it could
be automated) is not sufficient, and tasks that promote this skill usually require
to devise a non-trivial idea to address the problems they present.
The tasklet “Find the thief”, proposed by Belgium delegates in 2016 and
depicted in Fig. 1(c), is a good representative for this task since, in order to
estimate the number of interrogations required to detect the thief, students need
to understand that examining all visitors sequentially is too time consuming and
they need to address the problem with an original approach (in this case, binary
search). Thus, the task would be a good choice to promote the ability of devising
strategies to solve problems. Notice that, dealing the task also with complexity
issues, it could be used to promote the next skill, too.
2.6 Analyzing Algorithmic Solutions
This skill is promoted by tasks concerning global characteristics of the considered

algorithm, like correctness or complexity. Thus, tasks for this skill may require
to examine an algorithm (or, more generally, a computation method) in order
to understand its semantic, predict its overall behavior, determine its invariant
properties, estimate how many resources it will consume. Moreover, other typ-
ical tasks are those inspired by optimization problems, in that they require to
compare and evaluate different approaches in order to find the best one.
An example is given again by “Find the thief” (Fig. 1(c)): if a teacher wants
to introduce computational complexity, s/he can start from this task whose
solution relies on binary search and requires a complexity analysis in order to
estimate the right number of interrogations needed.
2.7 Implementing Algorithmic Solutions
Tasks that promote this skill may be referred to as programming or coding

tasks since the focus is on the implementation of algorithms according to a
formal syntax defined in the task. If the algorithm to be implemented is not

straightforward, then the task will also belong to skill Algorithmic thinking or
Identifying strategies.
Tasklet “Chain”, proposed by Slovak delegates in 2016 and depicted in
Fig. 1(d) is apt to teach this skill: a simple programming language is provided in
the text of the task and a small program has to be implemented to answer the
question. Actually, the multiple choice form of the question allows to find the
right answer by simply executing four programs; however, from a didactic point
of view, the tasks is perfect to introduce programming to young pupils.
3 Key Informant Interviews

To explore the suitability of using CT skills to present Bebras tasks and make
them easier to retrieve as a teaching resource, we used the key informant tech-
nique, i.e., we interviewed eight teachers of different kinds of schools, selecting
them for their first-hand knowledge, expertise and reflective practice in teaching.
Half of them were quite expert of Bebras tasks, in that they had participated in
the Bebras challenge in the past or had contributed in the preparation of tasks
themselves; the others instead knew Bebras only a little and are not expert of
computing education, but are very much concerned about curricular issues, pro-
fessional development of teachers, or teaching methodologies; some of them had
participated in our teacher training courses in computer science education.
Teachers were first provided with a short description [4] of the CT skills
presented in Sect. 2 and were then requested to associate some Bebras tasks
with those CT skills. More precisely, for each task and each CT skill, we asked
them the following question “Would you use this task to promote such skill?”,
emphasizing that each task could be associated with more skills. To express their
answers, we suggested that they fill out a double entry table (task/skill), where
the answers might range between 0 and 3: 0 = ‘absolutely not’; 1 = ‘more no than
yes’; 2 = ‘more yes than no’; 3 = ‘absolutely yes’.
We then (qualitatively) interviewed them and asked them to discuss their
answers with us, in order to understand their motivations. The interviews were
loosely structured, mainly relying on three wide issues.
– Are the (descriptions of the) skills clear? Are there terms or expressions that
you do not understand or about which you are not sure or that are ambiguous?
– Why do you relate a certain task to a certain skill? (In particular we delved
into those cases where the association between a task and a skill was unex-
pected for us).
– Do you think that if tasks were presented with this approach (that is consid-
ering the computational thinking skills they can promote), the educational
value of the tasks would emerge more clearly? That is, if presented this way,
would it be easier for a teacher to use them in the curricular educational
activity, as a curricular resource?
During the interviews, we first noticed that despite the effort of removing
(or reducing to the minimum) computer science technicalities, our description
of skills still revealed some computer science implied meanings that teachers
are not familiar with, and hence needed to be clarified, especially for teachers
of primary schools. For instance, when we use terms like problem and solution
we implicitly think of computational problems and solutions, while in primary
schools a problem is what we would call an instance of a problem, and vice versa
a computational problem would be seen as a class of similar problems in the
primary school meaning. Similarly, for primary school teachers with no formal
computer science education, a solution is simply an answer to an issue, while
when we write “analysing a solution” we are usually thinking of a computational
solution for a (computational) problem, i.e., an automatic method to find the
correct answer to any instance of the problem. Other expressions that ran into
a similar misunderstanding are representing data and organizing data. Indeed,
the digital representation of data as usually meant in computer science is not
common knowledge for primary school teachers. When a task deals with some
formal/symbolic representation of data, teachers realize that, to understand and
tackle the task, one needs to rearrange or reformulate the data somehow and
hence they associate the task with the “organizing data” skill. Also the terms
implement and coding were not broadly familiar and needed some explanation.
Despite the need to clarify these terms, in general we got confirmation of our
hypotheses. We agreed on most associations between tasks and CT skills the
teacher highlighted, which seems to confirm that the definitions of skills are
understandable and their use to describe tasks is feasible. Teachers seem to
appreciate the use of CT skills to analyze Bebras tasks since “make the underly-
ing skills of a task explicit helps in choosing more consciously what to work on
and how”. The CT skill lens seems to foster the identification of the educational
potential of tasks; indeed, often teacher highlighted associations between tasks
and skills that we did not expect, but they were usually able to support their
association with a convincing reasoning, or with clear examples (for instance
envisaging original ways to use the tasks in the classroom, in order to promote
a skill that was not directly addressed by the question of the task itself). From
the interview we also got some new ideas and insights. We noticed that tasks
that required some kind of analytic thought were often associated by teachers
with “algorithmic thinking”. For instance, solving a task by using the technique
of decomposition or step-by-step reasoning was often associated with algorithmic
thinking even when no algorithm or formal procedure or rule was involved. This
could be acceptable at the primary school level but, for older kids, algorithmic
and analytical thinking should be distinguished more neatly by teachers, in order
to appreciate the true computational thinking value of a task, and to promote
it in general education.
4 Categorization of Bebras Tasks Using CT Skills

Bebras tasks categorization is also important for the contest itself in that it helps
authors span on different topics/skills when they produce tasks and it helps
national organizers in putting up a varied contest covering as many informatics

topics and computational thinking skills as possible. In fact tasks categorization
is an issue in the Bebras community since the beginning. The classification pro-
posed in 2008 [7] turned out to be too coarse to be applied to a variety of tasks;
for example Table 1 shows how 120 tasks were classified in the 2016 edition (in
fact only 92 of them were indeed classified, since in the other cases the authors
of the tasks did not mark them, and we did not complete the classification).
Other categorizations based on the informatics content were proposed first
in [12] where four components of informatics education are considered (digital
literacy, programming, problem solving, and data handling) and more recently
in [13] where a hierarchical classification is suggested based on Schwill’s master
ideas [14] (algorithmization, structuring, formalization); a classification starting
from the Bloom Taxonomy of cognitive skills is presented in [10].
Finally, in [9] a two-dimensional classification is proposed: the first dimension
is based on informatics knowledge and proposes five informatics “domains”, the
second one is based on five CT skills. The five informatics domains, described
by means of several technical keywords (e.g., bubble sort, binary tree, etc.),
were also proposed and tentatively used to classify the 2017 tasks during their
creation and discussion, with the vast majority of tasks falling in the first two
categories. We also categorized 120 tasks from the 2016 edition within these
five domains; the results are given in Table 2. The second dimension, including
abstraction, algorithmic thinking, decomposition, evaluation, and generalisation
as CT skills, is indeed in the same direction we are proposing here. As such,
however, it is very high level, with some categories (especially abstraction) that
risk to be too ubiquitous to be useful. We chose not to apply such a classification
on our own, in order to avoid misinterpreting the original authors’ intention. We
instead applied our own scheme to classify the same 120 tasks from the 2016
edition, with the results shown in Table 3, also showing cases in which two (or
more) categories were used together, in bold the number of times a category was
used alone.
Table 1. Classification of 120 tasks from Table 2. Classification of 120 tasks from
2016 edition according to [7] 2016 edition according to [9]
Topics and concepts No. tasks Topics and concepts No. tasks
Algorithmic thinking 67 Algorithms and programming 87
Information comprehension 27 Data, data structures and 37
Structures, patterns and 16 representations
arrangements Computer processes and 15
Puzzles 16 hardware
Social, ethical, cultural, 1 Communication and 3
international, and legal issues networking
Using computer systems 3 Interactions, systems and 2
society
Table 3. Classification of 120 tasks from 2016 edition: number of tasks and category
pairs; bold figures count the occurrences of a category alone
CT skill No. tasks OD AD RD AT ID AS IS

Organizing data (OD) 17 4 8 2 3 0 0 0
Analyzing data (AD) 31 11 8 3 1 0 0
Representing data (RD) 25 8 0 0 7 0
Algorithmic thinking (AT) 40 19 0 5 9
Identifying strategies (ID) 11 5 5 0
Analyzing strategies (AS) 20 3 0
Implementing a solution (IS) 15 6
5 Conclusions
Bebras tasks are a considerable teaching resource, but unfortunately they are
mostly underused beyond the contest.
We analyse and describe Bebras tasks by using the operational definition of
computational thinking and identifying seven fundamental CT skills, concerning
the organization, analysis, representation of data, algorithmic thinking, and the
design, analysis and implementation of algorithmc methods.
Such an approach applies to a wide range of ages and schools and can be
described with terms and concepts that do not rely on a strong or wide knowl-
edge of computer science, hence also teachers without a formal education in
informatics can understand it and relate it with their curricular teaching activ-
ity. Moreover, differently from other classifications based on a taxonomy of infor-
matics topics, this approach also helps in detecting the cognitive skills involved
by a task, thus it would make their educational potential more explicit.
We gathered feedbacks about this approach by interviewing teachers and
applied it also to classify Bebras tasks. We will also collect feedbacks by building
a website where teachers will be able to retrieve tasks according to their potential
in promoting CT skills. We are currently working to link such approach to the
recommendations of the Italian Ministry of Education for non-vocational schools.
Finally, and overall, we believe the awareness of the importance of a CT-
based classification can improve the way tasks will be written in the future. In
particular we suggest to expand the Bebras task templates with a new “It’s
computational thinking” section, containing for each class an articulated answer
to the question “Is this task suited to teach this CT skill?”.
Acknowledgments. We would like to thank the whole Bebras community for the
great effort spent in producing exciting tasklets and the key informant teachers.
References
1. International Society for Technology in Education & Computer Science Teachers
Association: Operational definition of computational thinking for K-12 education
(2011). http://www.iste.org/docs/ct-documents/computational-thinking-operatio
nal-definition-flyer.pdf
2. International Society for Technology in Education & Computer Science Teachers
Association: Operational definition of computational thinking for K-12 education
(2017). http://www.csteachers.org/resource/resmgr/472.11CTTeacherResources
2ed.pdf
3. Lonati, V., Monga, M., Morpurgo, A., Torelli, M.: What’s the fun in informatics?
working to capture children and teachers into the pleasure of computing. In: Kalaš,
I., Mittermeir, R.T. (eds.) ISSEP 2011. LNCS, vol. 7013, pp. 213–224. Springer,
Heidelberg (2011). https://doi.org/10.1007/978-3-642-24722-4 19
4. Lonati, V., Malchiodi, D., Monga, M., Morpurgo, A.: Bebras as a teaching resource:
classifying the tasks corpus using computational thinking skills. In: Proceedings of
the 2017 ACM Conference on Innovation and Technology in Computer Science
Education, ITiCSE 2017, Bologna, Italy, 3–5 July 2017, p. 366. ACM (2017)
5. Corradini, I., Lodi, M., Nardelli, E.: Conceptions and misconceptions about com-
putational thinking among Italian primary school teachers. In: Proceedings of the
2017 ACM International Computing Education Research (ICER 2017), August
2017, to appear
6. Dagienė, V.: Supporting computer science education through competitions. In:
Proceedings of the 9th WCCE 2009, Education and Technology for a Better World,
Bento Goncalves (2009)
puter literacy: criteria for good tasks. In: Mittermeir, R.T., Syslo, M.M. (eds.)
org/10.1007/978-3-540-69924-8 2
8. Dagienė, V., Sentance, S.: It’s computational thinking! Bebras tasks in the cur-
riculum. In: Brodnik, A., Tort, F. (eds.) ISSEP 2016. LNCS, vol. 9973, pp. 28–39.
9. Dagienė, V., Sentance, S., Stupurienė, G.: Developing a two-dimensional cate-
(2017)
10. Dagienė, V., Stupuriene, G.: Informatics education based on solving attractive
tasks through a contest. In: Key Competencies in Informatics and ICT, KEYCIT
2014, pp. 97–115 (2015)
11. Haberman, B., Cohen, A., Dagienė, V.: The Beaver contest: attracting youngsters
to study computing. In: Proceedings of the 16th Annual Joint Conference on Inno-
vation and Technology in Computer Science Education, pp. 378–378. ACM (2011)
12. Kalos, I., Tomcsanyiova, M.: Students’ attitude to programming in modern infor-
matics. Informática na Educação: teoria & prática 12, 127–135 (2009)
13. Pohl, W., Hein, H.W.: Aspects of quality in the presentation of informatics chal-
lenge tasks. In: ISSEP 2015 local proceedings, pp. 21–22 (2015)
14. Schwill, A.: Fundamental ideas of computer science. Bull.-Eur. Assoc. Theor. Com-
put. Sci. 53, 274–295 (1994)
Country Reports
Introducing Programming and Digital
Competence in Swedish K-9 Education
Fredrik Heintz1,2,3,4(B) , Linda Mannila1,2,3,4 , Lars-Åke Nordén1,2,3,4 ,

Peter Parnes1,2,3,4 , and Björn Regnell1,2,3,4
1
Linköping University, Linköping, Sweden
fredrik.heintz@liu.se
2
Uppsala University, Uppsala, Sweden
3
Luleå University of Technology, Luleå, Sweden
4
Lund University, Lund, Sweden
Abstract. The role of computer science and IT in Swedish schools has

varied throughout the years. In fall 2014, the Swedish government gave
the National Agency for Education (Skolverket) the task of preparing a
proposal for K–9 education on how to better address the competences
required in a digitalized society. In June 2016, Skolverket handed over
a proposal introducing digital competence and programming as interdis-
ciplinary traits, also providing explicit formulations in subjects such as
mathematics (programming, algorithms and problem-solving), technol-
ogy (controlling physical artifacts) and social sciences (fostering aware
and critical citizens in a digital society). In March 2017, the government
approved the new curriculum, which needs to be implemented by fall
2018 at the latest. We present the new K–9 curriculum and put it in
a historical context. We also describe and analyze the process of devel-
oping the revised curriculum, and discuss some initiatives for how to
implement the changes.
1 Introduction
In recent years, we have witnessed an active discussion surrounding the role

of programming and computer science (CS) for everyone (see e.g., [9,16]). As a
result, an increasing number of countries have introduced or are in the process of
introducing CS in their school curriculum. For instance, in Europe, the majority
of countries (17 out of 21) taking part in a survey conducted by the Euro-
pean Schoolnet in 2015 reported doing so [1]. The way in which this is accom-
plished varies. Some countries focus on K–12 as a whole, whereas others primar-
ily address either K–9 or grades 10–12. Some countries have introduced CS as
a subject of its own (e.g. Computing in England [3]) while others have decided
to integrate it with other subjects, by for instance making programming an
interdisciplinary element throughout the curriculum (e.g. Finland [6]). A review
of models for introducing aspects of CS in K-12 education is presented in, for
instance, [7].
https://doi.org/10.1007/978-3-319-71483-7_10
118 F. Heintz et al.
The role of CS and information technology (IT) in school curricula has – in

general – varied over the years, placing focus on different areas, ranging from
using technology as a tool to learning how the computer works and how to use
it to create programs. This has also been the case in Sweden. In this paper, we
present a historical overview of technology and CS in Swedish K–12 curricula,
leading up to a revised curriculum being accepted in March 2017, introducing
digital competence and programming as interdisciplinary traits, also providing
explicit formulations in subjects such as mathematics (programming, algorithms,
and problem-solving), technology (controlling physical artifacts) and social stud-
ies (fostering aware and critical citizens in a digital society). We describe and
analyze the process of developing the new curriculum, and discuss some initia-
tives for how to implement the changes.
2 History of IT and CS in Swedish Schools

Already in the early 1970s, the Swedish government gave the agency responsible
for school curricula the task of investigating the possibilities of starting to teach
computer technology in Swedish schools. In the curriculum from 1980 (Lgr 80),
CS (datalära) was introduced in mathematics in grades 7–9, with a clear focus on
letting students learn about computers. Computers were also mentioned in nat-
ural and social studies, with emphasis being on understanding the consequences
the rapid development has on both individuals and the society as a whole.
All students need to be taught about the use of computers in our soci-
ety and about the fast development in the field. In particular, students
should realize that the computer is a technological tool, being controlled
by humans. (Lgr 80, p. 107, free translation from Swedish)
National programs were initiated to prepare students for living in a world

where computers were thought to play a significant role. The UIDA programme
(Utbildning inör datorsamhället, Education for the computer society) aimed at
giving all students in grades 7–9 “such knowledge that they can, want and dare
to influence the use of computers in the society” [10, p. 104]. In the mid 1980s
a specialized school computer called COMPIS (COMPuter in Shool) was intro-
duced. It rather quickly lost ground to the PC due to its limited functionality.
In the 1990s, computers became easier to use, the number of existing soft-
ware grew and the user interface became more easily approachable. In addition,
teachers rarely had sufficient competence for teaching CS, in particular program-
ming. As a consequence, focus was switched from CS towards digital literacy,
i.e. how to use applications, tools and computers. Instead of learning about com-
puters, the curriculum from 1994 (Lpo 94) focused on helping students learn
with computers. At the same time, there was a number of initiatives introducing
computers in school, for instance the DIG (Datorn i grundskolan, the computer
in school) and DOS (Datorn och skolan, the computer and school) projects.
Access to computers at homes also rose during the same time as the Home-PC-
reform was introduced. This reform made it possible for employees to acquire
Introducing Programming and Digital Competence 119
a computer from their employer against a gross deduction. In the early 2000s,
initiatives such as ITiS (IT i Skolan, IT at school) and PIM (Praktisk IT- och
Mediekompetens, practical IT and media competence) aimed at helping teachers
find good ways of using computers at school. The most recent curriculum reform
took place in 2011 (Lgr 11). For a more detailed historical overview of computers
and programming in Swedish education see [11].
3 Reintroducing CS in Swedish Basic Education

The Swedish school debate has, in recent years, circled around poor PISA results,
difficulties in providing all children and youth with equal opportunities, and
about modernizing the curriculum to meet changing job market requirements.
As “programmer” is one of the most common jobs in Stockholm, and the need
for software professionals is estimated to increase heavily both in Sweden and
internationally, some have argued that the educational system should teach pro-
gramming to prepare young people for these jobs. Others, the authors of this
paper included, believe that school should offer all students general prepara-
tion for any kind of work, and have therefore argued for digital competence as
part of all-round-learning, including computational thinking as a set of general
problem-solving skill useful for all in the spirit of Jeannette Wing [17].
In 2012, the Swedish government established the Digitalization Committee
(Digitaliseringskommisionen) with the task of providing guidelines for the future
of work related to digitalization in Sweden. One of the committee’s reports [4]
highlights the need for the school system to put larger focus on digital compe-
tence. The report explicitly points out the need for including programming in
the curriculum as part of existing subjects.
As a result of the discussion around schools, programming and CS as part of
all-round learning, persons representing school, universities and industry engaged
in voluntary initiatives to help bridge the lack of CS in Swedish basic education.
Teacherhack [14] is a nonprofit organization aiming at inspiring “teachers to
hack the current curriculum (Lgr 11) to include the essential skills that stu-
dents need in a digital world”. The Teacherhack website provides reviews of
all subjects in Lgr 11 with practical advice on how the current texts can be
interpreted in order to allow for a more active inclusion of content and prac-
tices related to CS, programming and the Internet, as well as to security and
integrity issues. Extracurricular activities such as CoderDojos, code camps, after-
school clubs and makerspace activities are organized to give children and youth
access to informal learning opportunities. Teachers throughout the country are
experimenting and sharing experiences from introducing programming in differ-
ent subjects, ranging from languages to handicraft and music. Heintz et. al. [8]
present an overview of ongoing activities related to CS and computational think-
ing in Sweden, highlighting several projects that show how one can introduce
CS already within the current curriculum.
In September 2015, the Swedish government gave the National Agency for
Education (Skolverket) the task of presenting a national IT strategy for the
Swedish school system. As one part of this work Skolverket was to update the
curricula for primary (K–9) and upper secondary education (grades 10–12). The
government explicitly stated that the curriculum should (1) strengthen students’
digital competence and (2) introduce programming at the K–9 level.
Vinnova, Sweden’s innovation agency, has also funded a number of research
projects related to the digitalization of school. One of these, the Trippel Helix
Project (www.trippelhelix.se), aims at getting school, industry and academia,
with regional backing, to together formulate a common concrete and achievable
action plan. The plan should be coordinated with the IT-strategy of the Swedish
National Agency for Education, pushing intellectual and operational changes
in the school of the future, based on the possibilities and knowledge demands
created by the digitalization.
4 Process
The directive from the government to Skolverket called for involving relevant
agencies, organizations and groups in the process. Skolverket formed several
reference groups, including one in CS education research. The authors of this
paper have participated in the process from two different perspectives; as a
reference group in CS education research and as part of the Trippel Helix project.
In the first stage, the researchers in the CS education reference group were
asked to review the current curriculum for grades 1–9 and suggest changes
and additions supporting the introduction of programming and digital compe-
tence. The reference group met with Skolverket in December 2015 and presented
their input. All researchers suggested that digital competence and programming
should be integrated in as many existing subjects as possibly (preferably all of
them), and not as a subject of its own. The role of CS was also emphasized,
rather than focus on programming which is seen as a particular skill within the
broader subject of CS.
Within the Trippel Helix project, representatives from industry, academia
and the school system work together collectively on supporting the digitalization
of education. During January and February 2016, three workshops were arranged
in different parts of Sweden (Stockholm, Gothenburg and Lund) to collect input
to the National IT Strategy and the new curriculum, both K–9 and 10–12. The
outcome of these workshops was handed as input to Skolverket’s ongoing work.
In early March 2016, a first draft including suggested curriculum changes
was sent out to different interest groups, with a request for feedback. Many of
the additions suggested by the CS education group were included in the draft,
but quite a few had also been left out. A few weeks later the first public draft
was published on the web. The authors provided feedback on the public draft
from a CS education perspective. In late April, the final public draft for a new
curriculum was published for anyone to give feedback on, with no explicit request
for feedback from the reference groups. In June 2016, Skolverket handed over a
proposal for revising the current curriculum to the government.
5 Revised Curriculum for Grades 1–9

In March 2017, the Swedish government accepted Skolverket’s proposal [15]
introducing a new general section on the importance of digital competence. The
new curriculum is mandatory from fall 2018, but schools and teachers are free
to start following it earlier if they want to.
Skolverket acknowledges that the meaning of digital competence changes over
time due to changes in society, technology and available services [12]. Skolver-
ket’s definition is based on the DigComp framework developed by the European
commission [5] and the work by Digitaliseringskommissionen [4]. In the Swedish
curriculum, digital competence includes four aspects: (1) understanding how the
digitalization affects individuals and society, (2) understanding and knowing how
to use digital tools and media, (3) critical and responsible usage of digital tools
and resources, and (4) being able to solve problems and implement ideas in prac-
tice. In a supplemental material [12], Skolverket stresses that the responsibility
of helping students develop their digital competence is a responsibility for all
subjects.
Programming is considered part of this definition. The supplemental material
clarifies that focus is not on coding skills, but on programming as a pedagogical
tool and problem-solving process including many phases. Programming should
also be seen in a wider context, including “creation, controlling and regulating,
simulations and democratic dimensions” [12, p. 10, freely translated]. Skolverket
emphasizes the importance of seeing programming in this wider perspective both
as a basis for teaching and as part of all four aspects of digital competence.
This provides the general setting for the revised curriculum and the four
aspects of digital competence can be found throughout the curriculum, both
in general parts and in the course plans for specific subjects. In the following,
we outline the most important changes made to the content of three subjects –
mathematics, technology and social studies – to address various aspects of digital
competence. The outline also shows the progression within different subjects,
based on the three curriculum levels (grades 1–3, grades 4–6, and grades 7–9).
The Swedish curriculum categorizes content for each subject under a number of
headlines in order to make the document more readable.
Programming is introduced mainly as part of the mathematics curriculum
under the topics “Algebra” and “Problem-solving”. The progression is as follows:
– Grades 1–3: How unambiguous step-wise instructions can be constructed,
described and followed as a basis for programming. The use of symbols for
step-wise instructions.
– Grades 4–6: How algorithms can be created and used for programming.
Programming in visual programming environments.
– Grades 7–9: How algorithms can be created and used for programming. Pro-
gramming in different programming environments. How algorithms can be
created, tested and iteratively improved when using programming for math-
ematical problem-solving.
The subject technology was introduced in the Swedish curriculum in 1994, with
the goal of helping students develop knowledge and skills needed for orienting
and acting in a technology intensive world. The revision puts increased attention
on digital technology and the need for developing an understanding for how
computers and networks work. The changes are made under three main topics.
For the topic “Technical solutions”, the revisions focus on understanding aspects:
– Grades 1–3: What computers are used for and some fundamental devices for
input, output and storage of information, for instance, keyboard, monitor and
hard drive. Some common artifacts that are controlled by computers.
– Grades 4–6: Some of the components and functions of a computer, for instance
processor and working memory. How computers are controlled by programs
and can be connected through networks.
– Grades 7–9: IT solutions for exchanging information, such as computers,
Internet, and mobile phones. Technical solutions that use electronics and how
these are programmed.
Under the topic “Development of technical solutions”, focus is placed on pro-

gramming as a means for controlling and creating technology.
– Grades 1–3: Controlling artifacts through programming.

– Grades 4–6: Controlling own constructions or other artifacts through pro-
gramming.
– Grades 7–9: Controlling and regulating own constructions, for instance using
programming. How digital tools can support the development of technical
solutions, for instance making drawings and doing simulations.
The related content under the topic “Technology, man, society and the environ-
ment” focuses on safety and integrity aspects:
– Grade 1–3: Safety when using technology, for instance when handling elec-
tricity and using Internet services.
– Grade 4–6: Safety when using technology, for instance when transmitting
information digitally. How technology is part of and changes the basic condi-
tions for different professions within all areas of society.
– Grade 7–9: Internet and other global technical systems, their benefits, risks,
and limitations. Safety when using technology, for instance storing and pro-
tecting data.
In social studies, the changes focus on digital competence, understanding the

influence that the digital transformation has on us as both individuals and a
society, and helping students become critical and responsible citizens in today’s
digitalized society. For the early grades (1–3), the previous texts on norms and
rules as well as evaluating sources have been expanded to also include digital
contexts. For grades 7–9, the main revisions concern two topics, with the pro-
gression for the topic “Information and Communication” being the following:
– Grades 4–6: How to act responsibly regarding the use of digital and other
media from a social, ethical, and legal perspective.
– Grades 7–9: The valuation of news and how it can be influenced by people’s
views of the world. How individuals and groups are portrayed, for example
based on gender and ethnicity as well as how information in digital media is
controlled through hidden programming. The opportunities and risks asso-
ciated with Internet and digital communication through electronic media as
well as how to act responsibly regarding the use of digital and other media
from a social, ethical, and legal perspective.
Under “Societal resources and distribution” the progression is as follows:

– Grades 4–6: The importance of digitalization for the individual, for example
in the form of improved possibilities for communication and e-commerce.
– Grades 7–9: The importance of digitalization for the development of society
in different areas, for example influence on the job market and infrastructure
as well as changed attitudes and values.
In addition to the quite notable revisions made in these subjects, other subjects
are also affected by the revision. Concrete examples are for instance the intro-
duction of modeling and simulations in natural sciences, extending the range of
materials and techniques to be used in crafts and emphasizing responsible com-
munication in languages. In summary, digital competence is visible throughout
the new curriculum, both in the general parts and in specific subjects.
6 Implementation
Revising the curriculum is only the first step, the next challenge is to implement
these changes in practice. There are more than 200 000 teachers in Sweden
and the need for in-service professional development is huge. The important
question is: How can teachers throughout the country be supported in teaching
programming and digital competence in their subjects?
Although the situation when introducing Computing in the curriculum in
England was somewhat different (own subject, ICT teachers already available at
schools), many of the challenges are the same. While ICT teachers were proficient
in teaching how to use technology, most of them had no prior experience in
programming or other topics focusing on understanding the technology. Hence,
one can expect to learn from their experiences in Sweden as well.
In England, companies such as Microsoft and BBC have supported the intro-
duction of Computing. The support organization Computing at School (CAS)
has helped build a network of teachers, who teach their peers, and a community
supporting teachers through local hubs, material distribution, and professional
development. It may seem as if teachers in England have received sufficient sup-
port. The transition to Computing has, nevertheless, not been unproblematic.
After the first year of having Computing, two main challenges were found [13]:
teachers’ limited programming skills and a lack of self-esteem for teaching the
new subject. Even though many of the teachers had taught ICT previously, most
of them had no prior experience in programming and CS. The formal professional
development offered has not been sufficient and many teachers have felt the need
to learn more on their own. Before the first school year of teaching Computing,
60% of all teachers felt they did not know enough to teach the new subject [2].
After the first term, almost half of the students felt teachers needed more train-
ing and over 40% of the students said that they had helped their teachers. Over
80% of teachers called for more training, in particular in programming.
In Sweden, Skolverket has developed online course modules to support teach-
ers in learning to teach the new content. These modules include, for instance,
school leadership in a digitalized world, programming as an interdisciplinary app-
roach, and programming in mathematics and technology. These modules can be
taken at any time, but the recommendation is for colleagues to get together
and take the course at the same time, forming collegial learning groups where
teachers can learn from each other and share experiences and ideas. In addi-
tion, Skolverket arranges conferences throughout the country aimed at showing
teachers what the changes mean in practice. Moreover, the supplemental mate-
rial describes what the new content entails and how it fits in the school context.
The universities are also in the process of slowly renewing the teacher train-
ing to meet the new requirements. Currently, the teacher training programs
do not include any mandatory courses in digital competence or programming.
Skolverket does not control university education, and can hence not place any
requirements on them. The universities do however need to educate teachers in
the competences and skills required by the current curriculum. Many universi-
ties are, for instance, planning courses on different aspects of digital competence,
including programming, aimed at both pre- and in-service teachers. One chal-
lenge is that the process of introducing new courses often have lead-times of
18–24 m. At least one university (Uppsala) has chosen to compensate for these
lead-times and pre-service teachers’ lack of training in digital competence and
programming by offering complementary education in the form of coding camps
offered to students that are graduating within the next 24 m. Other organiza-
tions, both public and private, also take an active role in making the transition as
smooth as possible. For instance, a school in Stockholm (Årstaskolan) has started
a portal of courses, which are freely available to anyone and help teachers learn
the basics of programming in different languages in a concrete manner. Swedish
national TV produces TV series, showing how to introduce programming to chil-
dren. Funding is offered for development and research projects, aimed at finding
good ways to teach the new content, hence contributing to a set of best prac-
tices. Dedicated social media groups gather teachers from all over the country
to share ideas and experience as well as answer each other’s questions. Regular
digichats are organized to provide an informal channel to discuss digitalization,
digital competence, and school.
The proposed IT strategy, which is not approved by the government yet,
also discusses important issues related to the new curriculum. For instance, the
need for updating the IT infrastructure. This includes access both to stable and
open Wi-Fi networks and computing devices (each teacher should have her own
device within two years and each student should have their own device within
three years). The IT strategy also discusses the need for computing devices to
be open for teachers and students to test and install their own software. This
is very important as in several cases in Sweden, the power of selecting learning
material has moved away from teachers and schools to their IT-departments.
That is, the format of the tool (from pen and paper to a digital device) dictates
who should decide on which learning material to use in the classroom. There are
examples where this process has led to computers and networks being totally
locked down with committees being the deciding body on which tools to allow.
The IT strategy also requires principals and other school leaders and admin-
istrative personnel to have an understanding of the issues involved. Principals
play a particularly important role in the implementation process as they must
promote and support their teachers in their teaching development efforts.
7 Discussion
In this section, we discuss the revisions from several perspectives including sub-
ject integration, bootstrapping, ambition, and equal opportunities.
Own subject or integrated approach. As noted above there seems to be two
ways in which countries are introducing CS or programming in their curriculum.
England provides a prime example of where the content has been packed into one
subject (Computing). In other countries, Sweden included, this has not been the
route taken; instead programming and CS content has been integrated in exist-
ing subjects. Introducing a subject of its own makes practical questions related
to who will teach the content unnecessary – dedicated Computing teachers teach
the subject Computing in England. The rationale for choosing an integrated app-
roach is manifold: (1) lack of space for introducing a new subject in the curricu-
lum, (2) letting students see and experience the use of programming in different
subjects (e.g. for raising interest among previously underrepresented groups),
and (3) introducing computational thinking, providing students a framework for
how they can work together with the computer to solve increasingly complex
problems. There are also practical considerations such as the negative experi-
ence of introducing the subject Technology in 1994 which needed about 20 years
to find its place, as well as the need for creating a teacher accreditation for a
new subject, which would take years to settle. In addition, as Sweden is only
revising the current curriculum, not creating a new one, introducing a totally
new subject would most likely not have been feasible.
Bootstrapping problems. The introduction of a new curriculum that spans
grade 1 to grade 9 naturally has to take into account a phased ramp up where
the progression is continuously adapted to the penetration of the curriculum
in terms of number of years of prior knowledge to build on. There is hence a
notable challenge in how to bootstrap the introduction of the revised curriculum.
Students in all grades (1–9) will need to start with the basics to be able to take
on more advanced topics later on. This requires teachers to take the curriculum
requirements for lower grade levels into account when designing the progression
for their current student group. A particular challenge arises in the intersection
between school levels, for instance, when students move from grade 6 to grade 7.
At this point many students move to another school, resulting in teachers having
a group of students, whose background in the new content can vary greatly
depending on which school they have attended and what teachers they have
had. The progression in the revised curriculum is expressed as if the curriculum
has already reached a steady state. This leaves local authorities throughout the
country with the responsibility of implementing the changes and solving the
bootstrapping challenge.
Too much or too little. As always when discussing changes and renewal
processes, critical voices are heard. Whereas some seem to think that schools
should not focus too much on digitalization aspects but pay more attention to
basic – and traditional – skills such as reading, writing, and calculating, others
think that the revisions are too vague and not bold enough. Developing steering
documents for education is difficult, as they should stay relevant for many years.
The situation is particularly difficult for a rapidly moving area such as infor-
mation technology. It is therefore crucial to focus on key principles and ideas
instead of detailed instructions and buzzwords. Describing content and learning
objectives in general terms is the only way to guarantee that a curriculum will
not be outdated after a short period of time. This also holds for the revisions,
which are written quite broadly, leaving ample room for interpretation. However,
these descriptions are often too vague for teachers and school leaders who are
to implement the curriculum in practice. There is hence a need for additional
material providing concrete examples of how the curricula requirements can be
implemented in the classroom. Such additional material can be updated and
expanded on in a fast and agile manner, compared to the curriculum. Expe-
rience and knowledge acquired from the large number of active projects and
collaboration initiatives between schools, municipalities, industry, and universi-
ties will be of great importance when implementing the changes at national level.
Material developed and lessons learned abroad can also help avoid reinventing
the wheel.
Equal opportunities and broadened participation. In the beginning, the
lack of detail in the curriculum can, however, be positive, as this leaves teachers
with a more easily approachable task. Instead of having detailed requirements,
they can do what they feel is enough and cover more content at more advanced
levels as they have learned more and become more experienced themselves. On
the other hand, this also introduces the obvious risk of teachers not raising
their ambition level as time goes by. Another risk is that all teachers will not
see the new content as their responsibility, but rather as part of some other
teacher’s duties. A challenge for the future is hence to get everyone to do enough
– this is a crucial aspect if we are to arrive at a school system providing equal
opportunities to all students regardless of where in the country they live and who
they happen to have as their teacher. The role of principals and school leaders
thus need to be stressed – it is their job to make sure that all teachers are able to
provide children and youth with the skills and content listed in the curriculum.
One important, but implicit, goal in the revised curriculum is to broaden the
participation in CS related study programs and jobs. This has also been one of
the important aspects that the authors of this paper raised in our discussions
with Skolverket. This is also one of the reasons why we argued for integrating
programming into as many subjects as possible, in particular also in aesthetic
subjects such as arts, handicraft, and music. The revisions, however, mainly
introduce programming in mathematics and technology, two subjects that can
be seen as the traditional and expected choice. We therefore see a risk that the
revised curriculum will not broaden participation as much as one would have
liked. At the same time, programming is introduced from grade 1 so everyone
will get exposed to programming from an early age, which may have a positive
influence on participation.
The role of Computer Science. The revised curriculum does not mention
computer science or any other synonym to the word. Rather, the focus is on
digital competence and programming. We believe this is very unfortunate, as
CS is the foundation for programming and includes much more than only pro-
gramming. This is further exacerbated by introducing programming mainly for
mathematical problem-solving and controlling technical artifacts. The interna-
tional trend, especially in the US, is to broaden the perspective and move the
focus from programming to the broader subject of computer science. This is also
related to the desire to broaden the participation as discussed above. To engage
and interest a broader group, the content needs to be broadened to speak to
this larger group. Further, CS is an academic subject, which programming is
not. To us introducing programming without CS is like introducing physical
experiments without introducing physics. We see upgrading CS to a status of
accepted general knowledge, similar to for instance mathematics and physics, as
an important goal.
8 Conclusions
Educational systems all over the world are being updated to reflect the possibili-
ties and challenges in the digitalized society. One part of these changes is related
to the inclusion of programming and computer science content in basic educa-
tion. The ways in which this is accomplished vary, and the Swedish revisions
are well in line with other countries, where the content is integrated in other
subjects.
Making a large-scale change in the education naturally takes time. The sit-
uation for students in different municipalities, schools, and even classrooms will
vary for many years before all teachers have had time to embrace digital compe-
tence and programming concepts as part of their everyday teaching. As the early
experience from England indicates, sufficient financial support is not enough;
what seems to be most important is access to continuous professional develop-
ment of high quality, suitable material, and peer support. In addition, school
leadership is of crucial importance, as teachers need time, access to professional
development, and support to make the best of the new situation.
Professional development is a continuous process, where teachers learn from

various sources and from each other. In the field of CS, the principles of life-long
learning become increasingly important, as technology moves forward at a fast
pace. Although the basic principles and ideas remain the same, new tools are
introduced at a regular basis, and teachers need to be able to evaluate and select
the right tool for their particular teaching situation.
The authors believe that the ongoing work on introducing programming and
digital competence in the curriculum for Swedish basic education is a good start,
although much still needs to be done. By presenting the new Swedish curriculum
and the practical considerations surrounding its implementation, we hope to add
to the discussion on how CS is being introduced in primary education throughout
the world. We think that an international discussion on teaching CS in basic
education is both important and valuable, both for sharing experiences and
learning from each other.
References
1. Balanskat, A., Engelhardt, K.: Computing our future. computer programming and
coding. priorities, school curricula and initiatives across Europe (2015)
2. Bateman, K.: Computing teachers need more training, say students, January 2015
3. Department for Education. National curriculum in England: Computing pro-
grammes of study (2013). https://www.gov.uk/government/publications/national-
curriculum-in-england-computing-programmes-of-study
4. Digitaliseringskommissionen. En digital agenda i människans tjänst : en ljusnande
framtid kan bli vår : delbetänkande. Technical report SOU 2014:13 (2014)
5. European Commission. DigComp 2.0: The digital competence framework for citi-
zens (2016). Accessed 22 April 2017
6. Finnish National Board of Education. Perusopetuksen opetussuunnitelman perus-
teet 2014 (2014)
7. Heintz, F., Mannila, L., Färnqvist, T.: A review of models for introducing com-
putational thinking, computer science and computing in k-12 education. In: 2016
IEEE Frontiers in Education Conference (FIE), pp. 1–9 (2016)
8. Heintz, F., Mannila, L., Nygårds, K., Parnes, P., Regnell, B.: Computing at school
in Sweden - experiences from introducing computer science within existing subjects.
In: Proceedings ISSEP (2015)
9. Informatics Europe and ACM Europe. Informatics in education: Europe cannot
afford to miss the boat, 2015. Report of the joint Informatics Europe and ACM
Europe Working Group on Informatics Education
10. Johanssen, M., Nissen, J.: It i framtidens samhälle och i dagens skola. Utbildning
och demokrati 10, 103–132 (2001)
11. Rolandsson, L., Skogh, I.-B.: Programming in school: look back to move forward.
Trans. Comput. Educ. 14(2), 12:1–12:25 (2014)
12. Skolverket. Få syn på digitaliseringen på grundskolnivå, June 2017
13. Swidenbank, R.: Coding in British schools: A review of the first term, January
2015
14. TeacherHack - En digitaliserad Lgr11. http://www.teacherhack.com/
15. Utbildningsdepartementet. Skolfs 2017:11 (2017)
16. White House. Computer science for all (2016). https://www.whitehouse.gov/blog/
2016/01/30/computer-science-all
17. Wing, J.: Computational thinking. Commun. ACM 49(3), 33–35 (2006)
Preparedness of Japan’s Elementary School
Teachers for the Introduction of Computer
Programming Education
Yutaro Ohashi(&)
Nippon Institute of Technology, Saitama, Japan

ohashi@nit.ac.jp
Abstract. By 2020, computer programming will be introduced to every

elementary school in Japan. This study explored the extent to which elementary
school teachers in Japan are prepared for this. A two-stage research plan was
adopted. The initial stage took a qualitative and inductive approach, in which the
author visited elementary schools to observe computer classes, interview
teachers, and administer a questionnaire. This stage was designed to identify
problems and note actual conditions in the field. This knowledge was used to
develop the questionnaire for the second survey. A quantitative and deductive
approach was used, involving a nationwide survey. Completed questionnaires
were returned by 309 elementary school teachers, covering 44 out of 47 of
Japan’s prefectures. The results show ICT to be currently underused in teaching
and revealed that many teachers lack confidence in, or are anxious about, the
plan. Four barriers were identified: the low level of ICT use in teaching, teacher
skepticism about the plan, the use of conventional teaching methods, and teacher
workloads. Overall, teachers were neither technically nor emotionally prepared
for the planned introduction of programming education. Ways of addressing
each barrier were proposed: increasing access to technological resources,
improving teacher training, introducing pupil-centered teaching and learning
methods, and reducing teacher workloads.
Keywords: Programming education Elementary school Teachers’

readiness ICT in education Japan
1 Introduction
1.1 Introducing Programming Education into Elementary Schools

in Japan
The introduction and integration of Information and Communication Technology
(ICT) into school curricula has presented major challenges internationally [16]. At the same
time, however, programming has been growing in significance in K-12 education. Its role in
the development of childhood skills has been acknowledged, and many countries have
incorporated it into their school programs, as programming, computing, coding, or com-
putational thinking [6]. Many model curricula are available [2, 4, 6, 10, 28].

https://doi.org/10.1007/978-3-319-71483-7_11
130 Y. Ohashi
Japan has specified that programming must be introduced into every elementary
school by 2020 [15]. In the Japanese context, “Programming Kyouiku” can be trans-
lated as “nurturing ‘programming thinking’, which will be required in every job when
the children become adults, by letting them experience information processing on
computers” [18]. While programming would be a critical part of such a curriculum, the
concept goes beyond coding. The specific course content, however, including the
programming language to be studied, the hours devoted to the subject, and the grade in
which the subject is to be introduced, is left to the individual school and teacher.
Programming will not be introduced as an independent subject, but should be taught in
combination with existing subjects [15].
Not surprisingly, there has been a debate over the feasibility of this approach.
Advocates argue that it is important to keep pace with other countries, where computer
programming is already compulsory at the elementary level [17], or that an under-
standing of coding is essential in a world in which digital equipment forms a large part of
daily life [8]. Opponents urge caution, and point to a shortage of teachers who are
qualified to teach coding [1]. A more specific concern is that the elementary level is too
early [25]. Despite the increase in interest, few studies have attempted to explore the
views of elementary school teachers on the planned reforms, or to examine the measures
that are being taken. Many case studies have examined advanced schools, and are of
little relevance to ordinary in-service teachers, especially given Japan’s relatively low
use of computers in learning activities [22, 23, 29]. This low usage has been attributed to
shortages of time, content availability, facilities, and resources available to teachers [9].
Many factors are known to play a role in determining ICT practices in the class-
room [5, 11, 12, 19, 24]. Key factors include access to resources, the quality of the
software and hardware available, the ease of use, the incentives for change, support and
collegiality, local and national policies, the commitment to professional learning, and
the amount of formal computer training that the teacher has undergone [20]. The beliefs
and attitudes of the teacher are also known to play a decisive role in programming
education, yet few studies have explored these.
1.2 Research Objectives and Research Questions

This study explored the attitudes towards and awareness of ICT use and programming
education by in-service elementary school teachers in Japan. The goal was to identify
potential and already existing obstacles and limitations. The following were the
research questions:
(1) How is ICT used by teachers in school affairs or teaching?
(2) What do teachers think about programming education?
2 Research Content
A two-stage approach was used. The initial stage was a qualitative and inductive, and
was used to gain an understanding of the problems and real conditions in the field, and
to establish the scope of the second stage. The author visited elementary schools to
Preparedness of Japan’s Elementary School Teachers 131
observe computer classes, interview teachers, and administer a questionnaire. The

second applied a deductive approach, based on a quantitative, nationwide survey and
reflecting the hypotheses arising from the first stage fieldwork.
2.1 First Stage: Pilot Study in Four Schools

The author visited four neighborhood elementary schools in Saitama Prefecture to
observe computer classes, interview teachers, and administer a questionnaire. The four
schools had applied to join our information volunteer program, in which students from
our university undertake activities such as supporting computer classes or preparing
teaching materials under the concept of service-learning [21]. Visits were arranged to
coincide with these activities, and lasted for about half a school day, allowing obser-
vations of two or three hours. The observer took notes, asked questions of the students,
and interviews the teachers, if possible. Teachers were then asked to complete a
paper-based questionnaire, to explore their use of ICT in school and their views on
programming education. In some schools, the questionnaire was administered by stu-
dents from the university.
The questionnaire contained no more than 10 question, excluding demographics
(age and years of employment as a teacher. Sex was not explored), and was restricted to
one side of an A4 sheet. Questions concerned ICT use in education and attitudes
towards programming education.
2.2 Second Survey: Nationwide Questionnaire

This pilot study was used to construct the main survey questionnaire. It was admin-
istered nationwide, with the help of a private online research service. Respondents were
recruited from monitors registered with the firm. The target response was not less than
300. To eliminate regional differences, sampling attempted to maximize prefectural
coverage.
2.3 Analysis Methods

In both surveys, multiple-choice questions were analyzed statistically and average
scores and correlations between responses were calculated. Responses to open-ended
questions were post-coded using a case-code matrix. This approach to analysis of
qualitative data is based on grounded theory [7], and identifies general patterns for a
specified case by fully considering both individuality and materiality [26]. Word
counting was not used in this study, as it does not always capture the significance of a
word and may decontextualize its description. The analytical procedure was as follows.
(1) Codes or keywords were assigned to each description using the following
sub-procedures:
(1-1) Open-ended coding to freely assign a code to a description
(1-2) Focused coding to assign a more abstract code to a description.
(2) Codes were structured to attain an overall balance.
(3) Steps (2) and (3) were repeated to ensure consistency across all codes.
132 Y. Ohashi
3 Results
3.1 First Survey: Fieldwork at School
The questionnaire contained 10 questions, and 19 responses were collected from six
schools. Questions Q1 to Q8 were multiple choice, and the responses are shown in
Table 1. Statistically significance at the one percent level were found for Q1 (ICT use
in school affairs), Q4 (confidence in using ICT), Q5 (familiarity with the proposals for
programming education), Q7 (experience of computer programming), and Q8 (confi-
dence in teaching computer programming). No significant correlation was found
between actual ICT use (Q1, Q2, Q3, and Q4) and confidence in teaching computer
programming (Q8). Responses to Q10 were coded into five categories (excluding “No
other comments” and other insignificant ones). Many of the comments reflected
uncertainty over the coming programming education (Table 2).
Table 1. Responses in the first survey (from Q1 to Q8).

N. Question Strongly Disagree Neutral Agree Strongly
disagree a little a little agree
Q1 Do you use ICT in 0 0 1 5 13
school affairs on a daily
basis?*
Q2 Do you use ICT in 2 1 4 4 8
teaching on a daily
basis?
Q3 Have you received 0 5 4 5 5
training to use ICT?
Q4 Are you confident in 1 5 4 9 0
using ICT?*
Q5 Do you know computer 2 1 1 3 12
programming will be
compulsory in
elementary schools?*
Q6 Have you received an 3 4 2 3 7
explanation of the
planned programming
education?
Q7 Have you received 10 3 1 3 2
training in computer
programming?*
Q8 Are you confident in 11 5 2 1 0
teaching computer
programming?*
* p < 0.01
Responses to Q9 included “support staff and/or help desk (16)”, “training and/or
instruction course for teachers (16)”, “new facilities such as computers and/or tablet
PCs (12)”, “time (10)”, “textbook (9)”, and “budget (8).”
Table 2. Responses in the first survey (Q10: Please write your opinions about ICT use or
programming education.)
Responses N.
Need for facilities and/or equipment 3
“If we have easy-to-use tablet PCs and other equipment we can teach pupils in all classes
in common (30s teacher)”
Anxiety 2
“I don’t know about programming at all. I’m anxious about it (40s teacher)”
Need for training and/or instruction course 1
“I need to serve apprenticeship for it (40s teacher)”
Need for specialized staff 1
“It would be helpful if specialized staff could come over and teach us (50s teacher)”
Expectation 1
“It is important for children. *snip* We’d like to explore the way we address it.” (50s
teacher)
Responses to Q10 were coded into five categories, including “need for facilities
and/or equipment (3)”, “anxiety (2)”, “need for training and/or instruction course (1)”,
“need for specialized staff (1)”, and “expectation (1).”
The results of the first survey are summarized as follows.
• Most respondents (18) reported using ICT in school affairs, but fewer in teaching
(12).
• Some respondents (15) reported knowing about the planned programming educa-
tion, but only half (10) had been informed about it.
• Measures concerning programming education had rarely been taken.
• Teachers felt uncertain about or dissatisfied with the planned programming edu-
cation. They felt under-informed and lacking the facilities and/or equipment needed,
or the chance to acquire the necessary knowledge or skills.
• A majority of respondents (16) expressed no confidence in teaching programming.
• Overall, the teachers surveyed were neither technically nor emotionally prepared for
programming education.
3.2 Second Survey: Nationwide Questionnaire

The first stage identified certain weaknesses in the questionnaire.
• It did not fully capture the views and attitudes of all teachers, including those who
are opposed to or anxious about programming education, and those who put it in
practice in advanced cases.
• Some terms were ambiguous. For example, it was unclear whether the option
“time” in Q9 referred to teacher time or a time slot in the curriculum.
To address this, the options in the questionnaire were revised. A company spe-
cializing in online research checked the consistency of wording and phrases. A na-
tionwide online survey was then conducted on August 3–5, 2017. Completed
134 Y. Ohashi
Table 3. Revised questions used in the nationwide questionnaire.

N. Question Answer
Q1 How many years have you been working as a Select from <10 to >40
teacher? years, by 10 year intervals
Q2 How often do you use ICT in school affairs? Six-point evaluation
Q3 How do you use ICT in school affairs? Multiple choice
Q4 How often do you use ICT in teaching? Six-point evaluation
Q5 If yes to Q4, how do you use ICT in teaching? Multiple choice
Q6 Are you confident in using ICT in school affairs Five-point evaluation
and/or teaching?
Q7 Do you know computer programming will become Yes/No
compulsory in elementary school (in Japan)?
Q8 Do you think programming education is necessary Five-point evaluation
in elementary school?
Q9 Have you taught pupils programming at school? Yes/No
Q10 If yes to Q9, what programming language or Multiple choice
development environment did you use?
Q11 Are you confident in teaching computer Five-point evaluation
programming?
Q12 What is (are) needed to teach programming at Multiple choice
school?
Q13 What do you expect from programming education? Multiple choice
Q14 Please write your any opinion about ICT use or Open ended
programming education.
questionnaires were returned by 309 respondents from 44 prefectures (out of 47). All
were elementary school teachers (Table 3).
• Demographics of the respondents (Q1)
Forty respondents (12.9%) were in their 20s, 66 (21.4%) in their 30s, 73 (23.6%) in
their 40s, 103 (33.3%) in their 50s, and 27 (8.7%) over 60. The average age was
44.9 years. The majority duration of employment was “less than 10 years” (104,
33.7%), followed by “10 to 19 years” (62, 20.1%), “20 to 29 years” (68, 22.0%),
“30 to 39 years” (68, 22.0%), and “more than 40 years (7, 2.3%).”
• ICT use in school affairs (Q2 and Q3)
Respondents used ICT in school affairs “almost every day” (198, 64.1%), “four to
five days a week” (38, 12.3%), “two to three days a week” (20, 6.5%), and “a day a
week” (10, 3.2%). “Rarely used” was reported by 32 respondents (10.4%) and
“never used” by 11 (3.6%). No statistically significant correlation was found
between age and years of teaching experience.
The reasons given for using ICT in school affairs were, in descending order,
“preparing for teaching” (252, 84.6%), “document preparation” (e.g. report) (243,
81.5%), “data management” (e.g. pupil scores) (255, 75.5%), “searching for
information” (219, 73.5%), “communication via mail and/or intranet” (144, 48.3%),
“crafting or making a presentation” (143, 48.0%)”, “miscellaneous work for club
activity” (27, 9.1%)”, and “other” (e.g. preparing a class paper, photo administra-
tion, etc.) (8, 2.7%).
• ICT use in teaching (Q4 and Q5)
ICT use in teaching was less common than in school affairs. Thirty seven
respondents used ICT for teaching “almost every day” (12.0%), 31 used it “four to
five days a week” (10.0%), 59 used it “two to three days a week” (19.1%), and 56
used it “a day a week” (18.1%)”. Meanwhile, 103 respondents reported using ICT
“rarely” (33.3%) and 23 answered “never” (7.4%), accounting for more than 40% of
respondents. The distribution was statistically significant at the one percent level.
No correlation was found with age or years of teaching experience (Q1).
The most frequently reported use of ICT use in teaching was “teacher shows
teaching material” (233, 81.5%), followed by “pupils search for information” (165,
57.7%), “pupils use educational software” (80, 28.0%), “pupils craft a presentation
of self-introduction or a survey conducted” (69, 24.1%), “pupils practice typing”
(60, 21%), “pupils practice programming” (9, 3.1%), and “others” (5, 1.7%).”
• Confidence in using ICT in school affairs or teaching (Q6)
When reporting their level of confidence, 40% responded “confident” (22, 7.1%) or
“somewhat confident” (104, 33.7%), 92 “neutral” (29.8%), 71 “not very confident”
(23.0%), and 20 answered “not at all confident” (6.5%).” The distribution was
statistically significant at the one percent level. A weak correlation was found with
the frequency of ICT use in school affairs (Q2) (r = 0.327) and frequency of use in
teaching (Q4) (r = 0.366). No correlation was found with age or years of teaching
experience (Q1).
• Awareness of programming education (from Q7 to Q13)
In total 247 respondents (79.9%) were aware that computer programming will
become compulsory.
Opinions about programming education varied. Eight respondents rated it “nec-
essary” (2.6%), 113 “somewhat necessary” (36.6%), 84 “neutral” (27.2%), 84 “not
very necessary (27.2%), and 20 “not necessary” (6.5%). The distribution was statis-
tically significant at the one percent level. No correlation was found with age or years
of teaching experience (Q1).
Forty five respondents (14.6%) reported already having taught programming at
school. Popular programming languages were Scratch (18, 40.0%), followed by Pro-
gramin (13, 28.9%), Viscuit (3, 6.7%), LEGO Mindstorm (2, 4.4%), Google Blockly
(2, 4.4%), Logo (2, 4.4%), and “other”.
More respondents were unconfident in teaching programming than were unconfi-
dent in using ICT, with a majority (65%) lacking confidence in teaching. Four
respondents answered “confident” (1.3%)”, 28 “somewhat confident” (9.1%), 76
“neutral” (24.6%), 107 “not very confident” (34.6%), and 94 “not at all confident”
(30.4%). A correlation was found with confidence in using ICT (Q6) (r = 0.608), but
not with age or years of teaching experience (Q1).
When asked what was needed to realize programming education, 221 responded
“support staff” (71.5%), 202 “textbooks, teaching material, supplemental teaching
136 Y. Ohashi
Table 4. Revised responses from the nationwide questionnaire.

Categories and examples N.
Need for fulfilling learning environment 87
“In any event, teachers are short on time. We have more and more tasks (49-year-old
male teacher)”
“ICT facilities vary depending on the local government. I’m anxious about programming
because we have one computer for two pupils. I cannot do what I want to do because
computers are too slow and internet filtering is too strict (32-year-old female teacher)”
Disbelief 81
“The local government where I work is in extreme poverty and barely in financial
reconstruction. Even cracks or leaky roofs cannot be repaired. *snip* I’m in doubt if the
ICT-driven classroom can contribute to develop pupils’ academic ability. Before
programming education, we should be there for pupils (41-year-old male teacher)”
“I have no idea in which subject we do programming. If it is left solely in the hands of
each teacher or school we don’t need it, as we already have other planned subjects such
as Ethics and English. The ministry should scrutinize what pupils learn (52-year-old male
teacher)”
Expectation 4
“I came to think the programming education is required because, for example, foundation
of English is nurtured in childhood. I learned electronic engineering with kits in my
childhood and it affected me (53-year-old male teacher)”
“I think it is important to nurture logical thinking. We need to consider ways of literacy
to address various social problems at the national level. (56-year-old male teacher)”
material” (65.4%) or “instruction course and/or training for teachers” (65.4%), 162
“time for research and/or practice” (52.4%), 129 “computers” (41.7%), 128 “tablet
PCs” (41.4%), 115 “Wireless LAN” (37.2%), 105 “a portal site for programming
education” (34.0%), 59 “helpdesk for customer call and/or e-mail” (19.1%), and
“others” (9, 2.9%). Sixteen respondents answered “nothing in particular” (5.2%).”
Most respondents (253, 81.8%) had positive expectations of programming educa-
tion: “knowledge and/or skills needed in the future” (132, 42.7%), “learning with joy”
(119, 38.5%), “logical thinking” (102, 33.0%), “creativity” (83, 26.9%), “catalyst to
think about future jobs” (80, 25.9%), “experiencing latest technologies” (66, 21.4%),
“mathematical thinking” (19.4%), 45 “teachers use computers (60, 14.6%)”, “coop-
erating with other subjects” (40, 12.9%)”, and “pupils come to like school and/or
studying through programming” (34, 11.0%).” Fifty seven respondents reported
expecting nothing (18.4%).
• Opinions about ICT use or programming education (Q14)
After excluding uninterpretable responses (e.g. “No other comments.”), 172 com-
ments were collected. Comments were analyzed using the method noted above and
divided into three categories: “need for fulfilling learning environment”, “disbelief”,
and “expectation” (Table 4).
4 Discussion
4.1 Revisiting Research Questions
We first discuss the two research questions noted in Sect. 1.2.
(1) How do teachers use ICT in school affairs or teaching?
Respondents have used ICT in school affairs and were reasonably confident in so
using it. However, ICT is underused in teaching. Even those who do use ICT in
teaching use it in a teacher-centered way (e.g. presentation of teaching material).
This confirms previous studies. A report by OECD noted that only 10% of
teachers in Japan reported frequent or regular use of ICT for student projects or
class work in all or nearly all lessons [22]. An OECD PISA country note stated
that computers have less presence in Japanese schools than the average across
OECD countries [23]. Cuban (2001) claimed that computers were oversold and
underused in US classrooms [3], reflecting the situation in classrooms in Japan.
(2) What do teachers think about programming education?
More than 80% of respondents had positive expectations of programming edu-
cation and 14% had already taught programming. However, more than half
expressed disbelief in or anxiety about the plan. Sixty five percent lacked confi-
dence in teaching programming, twice as many as lacked confidence in using ICT
in school affairs or in teaching (29.5%). Supporters of programming education
(39.2%) were only slightly more numerous than opponents (33.7%), suggesting
that this is a controversial issue even among teachers. In short, the teachers studied
were neither technically nor emotionally prepared for programming education.
4.2 Four Barriers and Provisions

Four barriers found from the two surveys, and possible remedial measures, are sum-
marized below.
(1) Low level of ICT use in teaching
The current underuse of ICT in Japanese classrooms is a potential major barrier to
its wider implementation. As Kusano et al. (2013) suggested, greater access to a
technological resource stimulates greater usage [12]. Japanese classrooms should
therefore prioritize increasing access. The correlation between confidence in using
ICT and confidence in teaching programming suggests that promoting ICT use
among teachers may extend its implementation. A realistic approach should reflect
our results; computer science unplugged could be a good option for those who
cannot use ICT well.
(2) Teachers’ lack of confidence in the plan
One notable finding of this study was the number of Japanese teachers who lack
confidence in or are anxious about the plan. Previous studies had not identified
this. The major reasons are that the teachers are poorly informed, or think the plan
fails to reflect actual conditions. The decision to use ICT in the classroom is
affected by the teacher’s commitment to professional learning and background in
formal computer training [20]. Proper teacher training is therefore essential. When
138 Y. Ohashi
designing a training plan, known barriers such as isolation, lack of adequate

computer science background, and limited resources for professional development
[30] should be taken into account. However, training more than four hundred
thousand in-service elementary school teachers [14] within the available time will
be a considerable challenge.
(3) Conventional teaching methods
In the course of the fieldwork, conventional teaching methods were observed in
the classroom. Some were teacher-centered even when computers were used, for
example all pupils doing the same exercise, or merely copying the teacher.
Paper-based teaching and learning styles were also encountered, even in computer
classes, with an emphasis on writing by hand. This may reflect Japanese writing
culture, which encourages many (and especially teachers) to believe that good
writing must be stressed in childhood.
While such conventional methods have a role in the classroom, they may also
construct an invisible barrier to ICT integration, which requires self-motivated
learning or learning by doing. The current discussion is mainly focused on coding.
An equal focus should be placed on the teaching and learning methods that are
best suited to programming education.
(4) Overworked teachers
The fourth barrier is the lack of teacher free time, which was found in the obser-
vational fieldwork rather than the survey. Teachers were working busily and were
heard to complain about overwork on several occasions. One female teacher
observed that she was unable to handle her current workload, and that the addition of
programming education would be impossible. This is supported by the fact that
Japanese teachers work the longest hours in the OECD member countries, due to the
extra load of administrative work and extracurricular activities (e.g. coaching clubs)
[27]. While measures are being taken to tackle this [13], overtime work is a custom
that remains rooted in Japan’s working culture. Ways must be found to lighten
teacher workloads. Our fieldwork observations and survey results suggest that this
will be a significant barrier to the successful introduction of programming education.
5 Conclusions
This paper explored the extent to which elementary school teachers in Japan are prepared
for the planned introduction of programming education. Field observations and a
nationwide survey were conducted. Completed questionnaires were returned by 309 ele-
mentary school teachers, covering 44 out of 47 prefectures. The results show that ICT is
currently underused in teaching and that many teachers lack confidence in, or are anxious
about, the plan. Four barriers were identified: the low level of ICT use in teaching, teacher
skepticism about the plan, the use of conventional teaching methods, and teacher work-
loads. Overall, teachers were neither technically nor emotionally prepared for the planned
introduction of programming education. Ways of addressing each barrier were proposed:
increasing access to technological resources, improving teacher training, introducing
pupil-centered teaching and learning methods, and reducing teacher workloads.
Acknowledgments. Author thanks all of the concerned parties, i.e., teachers and staff at the
participating schools, participating students, supporters of this study, and those who proofread
our work.
References
1. AERA: Few teachers can teach computer programming even though it will be compulsory in
elementary school from 2020 (in Japanese). https://dot.asahi.com/aera/2016102500219.html.
Accessed 10 June 2017
2. Computing At School (CAS): Computing in the national curriculum: a guide for primary
teachers (2012). http://www.computingatschool.org.uk/data/uploads/CASPrimaryComputing.
pdf. Accessed 10 June 2017
3. Cuban, L.: Oversold and Underused: Computers in the Classroom. Harvard University Press,
Cambridge (2001)
4. Du, J., Wimmer, H., Rada, R.: “Hour of Code”: can it change students’ attitudes toward
programming? J. Inf. Technol. Educ. Innov. Pract. 15, 52–73 (2016)
5. Ertmer, P.A., Ottenbreit-Leftwich, A.T., Sadik, O., Sendurur, E., Sendurur, P.: Teacher
beliefs and technology integration practice: a critical relationship. Comput. Educ. 59, 423–
435 (2012)
6. European Schoolnet: Computing our future: Computer programming and coding Priorities,
school curricula and initiatives across Europe (2015). http://www.eun.org/publications/
detail?publicationID=661. Accessed 10 June 2017
7. Glaser, B.G., Strauss, A.L.: The Discovery of Grounded Theory: Strategies for Qualitative
Research. Transaction Publishers, Piscataway (1967)
8. Harada, Y.: Peta-gogy for future: programming for children with viscuit. Inform. Process.
53, 60–64 (2011). (in Japanese)
9. Japan Association for Promotion of Educational Technology (JAPET): Actual condition
survey on ICT use in school (in Japanese). http://www2.japet.or.jp/ict-chosa/ict_chosa_data.
pdf. Accessed 10 June 2017
10. k12.org. https://k12cs.org/. Accessed 10 June 2017
11. Kim, C., Kim, M.K., Lee, C., Spector, J.M., DeMeester, K.: Teacher beliefs and technology
integration. Teaching Teacher Educ. 29, 76–85 (2013)
12. Kusano, K., Frederiksen, S., Jones, L., Kobayashi, M., Mukoyama, Y., Yamagishi, T.,
Sadaki, K., Ishizuka, H.: The effects of ICT environment on teachers’ attitudes and
technology integration in Japan and the U.S. J. Inf. Technol. Educ. Innov. Pract. 12, 29–41
(2013)
13. Ministry of Education, Culture, Sports, Science and Technology – Japan (MEXT): Making
school working environment proper (2016) (in Japanese). http://www.mext.go.jp/a_menu/
shotou/uneishien/detail/__icsFiles/afieldfile/2016/06/13/1372315_03_1.pdf. Accessed 12
Sept 2017
14. Ministry of Education, Culture, Sports, Science and Technology – Japan (MEXT): MEXT
Statistical Directory (2016) (in Japanese). http://www.mext.go.jp/b_menu/toukei/002/002b/
1368900.htm. Accessed 12 Sept 2017
15. Ministry of Education, Culture, Sports, Science and Technology – Japan (MEXT): A
concept of programming education at elementary school level (2016) (in Japanese). http://
www.mext.go.jp/b_menu/shingi/chousa/shotou/122/attach/1372525.htm. Accessed 12 Sept
2017
140 Y. Ohashi
16. Ministry of Education, Culture, Sports, Science and Technology – Japan (MEXT).
Informatization of education. http://www.mext.go.jp/b_menu/hakusho/html/hpac200101/
hpac200101_2_067.html. Accessed 25 August 2017
17. Ministry of Education, Culture, Sports, Science and Technology – Japan (MEXT): Report on
programming education in other countries (in Japanese). http://jouhouka.mext.go.jp/school/
pdf/programming_syogaikoku_houkokusyo.pdf. Accessed 10 June 2017
18. Ministry of Education, Culture, Sports, Science and Technology – Japan (MEXT): The
concept of programming education in elementary school (in Japanese). http://www.mext.go.
jp/b_menu/shingi/chousa/shotou/122/attach/1372525.htm. Accessed 10 June 2017
19. Msila, V.: Teacher readiness and information and communications technology (ICT) use in
classrooms: a South African case study. Creative Educ. 6, 1973–1981 (2015)
20. Mumtaz, S.: Factors affecting teachers’ use of information and communications technology:
a review of the literature. J. Inf. Technol. Teacher Educ. 9(3), 319–341 (2000)
21. Nejmeh, B.A.: Service-Learning in the Computer and Information Sciences: Practical
Applications in Engineering Education. Wiley, Hoboken (2012)
22. OECD: Result from TALIS (2013). https://www.oecd.org/japan/TALIS-2013-country-note-
Japan.pdf. Accessed 10 June 2017
23. OECD: PISA Students, Computers and Learning: Making the Connection, Country note
Japan. https://www.oecd.org/pisa/keyfindings/PISA-2012-students-computers-japan.pdf.
24. Prestridge, S.: The beliefs behind the teacher that influences their ICT practices. Comput.
Educ. 58(1), 449–458 (2012)
25. Sankei Shimbun. http://www.sankei.com/column/news/160801/clm1608010006-n1.html.
26. Sato, I.: Qualitative Data Analysis Methods, Shinyosha (2011). (in Japanese)
27. The Japan Times: Japanese teachers work longest hours among OECD members. http://
www.japantimes.co.jp/news/2014/06/26/national/japanese-teachers-work-longest-hours-
among-oecd-members/#.WTuDBJLyiCg. Accessed 10 June 2017
28. Tucker, A. (ed.): A Model Curriculum for K-12 Computer Science: Final Report of the ACM
K-12 Task Force Curriculum Committee (2003). https://www.acm.org/education/curric_
vols/k12final1022.pdf. Accessed 10 June 2017
29. UNESCO: Information and Communication Technology (ICT) in Education in Asia: a
comparative analysis of ICT integration and e-readiness in schools across Asia. http://www.
uis.unesco.org/Communication/Documents/ICT-asia-en.pdf. Accessed 10 June 2017
30. Yadav, A., Gretter, S., Hambrusch, S., Sands, P.: Expanding computer science education in
schools: understanding teacher experiences and challenges. Comput. Sci. Educ. 26(4), 235–
354 (2016)
Baltic and Nordic K-12 Teacher Perspectives
on Computational Thinking and Computing
Arnold Pears1(B) , Valentina Dagiene2 , and Egle Jasute2

1
Royal Institute of Technology (KTH), Stockholm, Sweden
pears@kth.se
2
Vilnius University Institute of Mathematics and Informatics,
Akademijos Street 4, LT-08663 Vilnius, Lithuania
{valentina.dagiene,egle.jasute}@mii.vu.lt
Abstract. This paper reports on the results of a study of teacher pre-

paredness and practices in relation to teaching computing and computa-
tional thinking at schools in Sweden, Finland and Lithuania. The study
was conducted as part of a NordForsk funded project to explore how
Computing Education Research in the Universities can help the devel-
opment of teacher training and K-12 curriculum and teaching practices.
The study found that many teachers are already engaged in teaching
relevant material in the schools, and that many have good support in
their local school environment. However, there are also significant chal-
lenges which emerge from the new curricula that have been introduced in
Sweden and Finland. To meet these challenges new teacher training pro-
grammes will be needed, and we recommend that computational thinking
and computing concepts be introduced into the core subject content of
teacher education programmes in order to better prepare teachers to
meet the educational demands of our increasingly digitalised society.
Keywords: Computational thinking · Teacher training · School cur-

riculum
1 Introduction
This paper draws on data collected over a two year period as a part of a two year
transverse action, the NordPlus Network on Innovative Computing Education -
NordNICE, funded by NordPLUS. The goal of the action was to increase the
interaction and knowledge flow between the compulsory schooling sector and
university research centers to enhance mutual understanding of how best to equip
future generations with the computational thinking and computing knowledge
which underpin the success of our increasingly technological society.
The network activities were designed to facilitate extensive collaboration
between educators, teachers, and researchers in computing teacher education,
both in pre-service and in-service training. Coordinated by Vilnius University,
the network was a collaboration between several Nordic and Baltic universities -
Uppsala University, Åbo Akademi University, University of Tartu, University
https://doi.org/10.1007/978-3-319-71483-7_12
142 A. Pears et al.
of Eastern Finland, Aalto University, KTH University, and Latvian University.

These universities shared expertise and collaboratively developed materials for
teaching computing together with partners in the schools of the participating
countries. The initiative introduced new methods of teaching programming,
using visualization tools in teaching, and using modern and innovative teach-
ing methods in teaching computer science to school teachers. The work was
conducted as a mix of live and online meetings to establish and develop the
Network. A programme of activities was carried out to develop teacher training
guidelines and a framework for computing curricula for schools. The initiative
developed mechanisms for sharing best practice, inclusion of new partners, and
dissemination of results. The ultimate goal is to improve the quality of com-
puting teacher education in all Nordic and Baltic countries, and in this way
support the development of top-level experts for the ICT industry and public
organizations.
Data from the NordNICE action included responses to a survey designed to
benchmark teacher perceptions of the relevance of computational thinking and
their associations with computational thinking and computing in general. While
the term Computational Thinking (CT) has been ascribed to Papert [1], many of
the fundamental ideas have existed for a long time [2]. The recent wave of change
in school education was sparked by the use of the term by Jeannette Wing in
her 2006 article [3] in which she suggested that computational thinking was “...
a fundamental skill for everyone, not just for computer scientists. To reading,
writing, and arithmetic, we should add computational thinking to every child’s
analytical ability.” [3] She followed this up with a more concrete definition in
2011 where she formulated CT as “the thought processes involved in formulating
problems and their solutions so that the solutions are represented in a form
that can be carried out by an information-processing agent” [4]. Crucial to this
latter definition is the idea that there are layers of abstraction of both data and
processes involved in computational thinking.
However, as has been demonstrated by the range of initiatives in the Baltic
region and internationally this is not a trivial process. Developing curricula
that integrate CT components into a broad range of school subjects is a long
term process requiring considerable systematic work. New curricula have been
devised and deployed [5–9], but, how well do teachers feel prepared to teach
these new curricula? One of the reported weaknesses of prior initiatives which
have introduced CT and computing content into the compulsory school curricu-
lum is the lack of attention to supporting teachers [10]. This general challenge
can be decomposed into several sub-components which need to be addressed.
The most critical to short term success is provisioning of in-service training for
teachers who are expected to implement new content in the contexts of their
school teaching. In the longer term there is also a need to revise the curricula
and courses in teacher education programs at universities to include new mate-
rial and to discuss how to deal with the specific pedagogical challenges that the
new curricula content presents for pupils.
Baltic and Nordic K-12 Teacher Perspectives 143
A major area of need is primary school teacher training, since these teach-
ers do not traditionally have any background in computing. To help inform the
development of new in-service teacher training initiatives the NordNICE part-
ners surveyed school teachers in 2015 and 2017 to gather perceptions of how
development of their competence in these areas was supported by their local
environment.
2 Status of Computing in School Curriculum

2.1 Lithuania
In Lithuania Computing (and CS) is referred to as Informatics (“informatika”).
As early as 1978–1979, proposals were developed for pupils’ education in pro-
gramming. In 1986–1997 Lithuania established compulsory informatics content
in all secondary schools. The course was lectured for two years with 70 h overall
in grades 11 and 12 (leaving certificate). The curriculum covered three main
areas: (1) the concept of information, principles of its transmission, storage and
processing; (2) logical principles of computers; (3) algorithms, particularly data
types and basic control structures of programming.
As a part of the education reform in 1997, the Informatics core curriculum
went through a major revision and it was expanded from two years to four
years teaching time (in total 136 h) with more focus on application and devel-
oping algorithmic thinking. The teachers were formally qualified, usually with a
bachelor or masters degree in informatics combined with mathematics. Teacher
training institutions offered various courses in programming and applications.
Since 2005, the main part of the compulsory informatics course has been
dedicated to computer literacy and applications. As a result, the teaching of
informatics has become quite poor. Pupils are introduced to the basics of infor-
matics in grades 5 and 6 based on Logo or Scratch [11]. For grades 9 and 10, an
optional module on algorithm design and coding is recommended for advanced
students. Consequently lower secondary school introduces basic informatics and
serves as a starting point for the informatics education of all students in high
school.
The most significant influence to deepen informatics education are shaped by
the development of the informatics maturity exam in 1995. Those, who pass the
informatics exam, have enhanced opportunities to enter CS-related studies in
higher education. The test provides a reliable indication as to whether a student
has the aptitude to study informatics. The informatics exam consists of two
parts: one part (over 50%) is allocated to programming, while the rest concerns
the issues of computer literacy and application. The exam focuses on: knowledge
and understanding 30%, skills 30% and problem solving 40%. Tasks are oriented
towards knowledge of data structures and the development of simple algorithms.
In 2016 the Lithuanian Ministry of Education and Science developed new
guidelines (framework) of new informatics curricula for grades 1–10 with special
focus on informatics concepts and skills [5]. Students should gain familiarity with
basic concepts in informatics in grades 5 and 6 when they learn coding with Logo
144 A. Pears et al.
and Scratch, etc. In grades 9 and 10 when they progress towards developing
algorithms. Education quality depends on teacher’s knowledge and engagement.
Outreach activities have a significant influence on informatics curricula e.g. the
Bebras Challenge on Informatics and Computational Thinking [6]. Students in
upper secondary school (grades 11 and 12) have the option of choosing advanced
modules on programming, data bases and a range of other CS topics.
2.2 Finland
In Finland, Computer Science (CS) was part of the upper secondary curriculum
(grades 10 to 12) until the 1994 national curriculum reform, later ICT (Informa-
tion and Communication Technologies) were integrated in all subjects. In 2016,
Finland introduced a new national curriculum for general education (grades
1–9). The new curriculum includes both ICT-related skills and programming.
Programming is explicitly included in mathematics education, starting with pri-
mary students giving instructions to each other and moving towards graphi-
cal programming environments and using programming languages in secondary
school. The introduction of programming in the core curriculum has resulted
in different initiatives facilitating and supporting this reform. For instance, a
teacher guide called Koodi2016 (Code 2016) was published in early June 2014
with both state and industry support [12].
The National Board of Education, which is in charge of the curricula reform,
provides some funding for professional development aimed at in-service teachers
as well as curriculum related to the use of programming environments and tools
in schools. However it is the duty of education providers like municipalities to
ensure a sufficient in-service training for teachers to increase the professional
skills and professional self-esteem.
The new curriculum has clearly intensified the use of computing in the classes
and classrooms. Many teachers are however looking forward to have a more
radical integration process to intensify the use computers in everyday situations
to make computers an essential part of learning.
The teacher students at the universities are introduced to computational
thinking quite deeply but a lot of practice in schools is needed.
2.3 Sweden
The first curriculum for K-12 CS education in Swedish schools was established
in the 1970s, under the term informatics, later changed to information tech-
nology. Informatics was a supplementary subject in vocational education, and
later became obligatory for upper secondary students in the natural science
programmes [9]. For a number of years computing related material was offered
at primary and lower secondary level, but was later removed as primary and
secondary teachers were not properly qualified to teach the content, and more
specifically the programming element proved hard for many teachers to teach.
Over the last decade computer science has transformed into a subject mainly
focusing on digital literacy at the primary and secondary levels, with primary
focus on applications and tools (e.g. the Office suite) and its applications in
other subjects. In Sweden’s compulsory school CS has not been taught as a
separate subject. Instead the content is offered as parts of other school subjects
like Technology and Mathematics. In the upper secondary schools (years 10–12),
the CS education varies depending on the programme attended.
Teacher training in Sweden does not include courses in CS, making it is
unusual to find teachers with computing expertise in many schools. There are
courses at upper secondary school in programming, multimedia, web design etc.
that are offered to any student who wishes to take them. However, programming
courses are mainly attended by students who attend one of the three programmes
in technology, natural sciences or electronics. Hence, historically only a minority
of students have taken programming courses.
The ICT content in the Swedish educational system is summarized as follows:
One ICT learning outcome is specified for Swedish preschools, which prescribes
that pupils shall become capable in the use of “multimedia and information
technology in creative processes as well as in applications.”
In the compulsory school (grades 1–9) ICT training focuses on equipping
students with the “how of information”, how to use digital technology, how
critical thinking can be applied to the information available on the Internet, etc.
In 2012, the Swedish government established the Committee for Digitization
(Swedish: Digitaliseringskommisionen) to provide direction and guidelines for
future work on the digitalisation of Swedish society. The committee recently
published a report which pinpoints the need for school to embrace the concept
of digital competency, in accordance with the DigiComp framework developed
in the European Union.
In 2017 the government introduced a revised curriculum which focused more
on digital competences and the development of relevant computing and compu-
tational thinking skills at all levels of the K-12 school system.
3 Teacher Attitudes to Computing
Programming and system development skills have been an issue in terms of

teachers’ competencies from the initial phases of all the initiatives we described
above. Riis [13] notes that “The teachers whose ideas are expressed in the school
reports are enthusiastic, competent, and generally very optimistic about the
future, and they have a constructive attitude toward computer use in school.
However, they hardly represent the majority of Swedish teachers today. Their
experience perhaps can not in all cases and without difficulty be transferred to
others. This is an area in which in-service training could give these teachers a
chance to share their experiences with those who have not been in the front
ranks.” To assess the current situation in the teaching community this study
of teacher attitudes to teaching computing in schools collected data using two
surveys (Survey I and Survey II) in the NordNICE project.
Survey I was developed during a NordNICE workshop in Druskininkai in
Lithuania in December of 2015. The full survey can be found in through the
146 A. Pears et al.
WWW link in Appendix A. This survey uses items from several earlier research
initiatives [14] and comprises three major sections, background demographics,
common classroom activities, and a concept association table.
The aim of the first survey was to obtain a broad picture of the attitudes
and associations of teachers in relation to central concepts in computational
thinking and computing identified in prior research. The survey was broadly
distributed through teacher networks and social media to teachers in all the
countries participating in the action.
Survey II uses a purposeful sampling approach with the intention to select
responses which represent a heterogeneous (maximum variation) sample. This
was done by intentionally selecting 3–5 schools in each country using “engage-
ment with computing” as a primary variable for selection. The goal was to collect
data from teachers who are already actively teaching computing content as well
as teachers in schools where teaching of computing content was not yet imple-
mented. Schools in Lithuania, Latvia, Estonia, Sweden and Finland participated
in this second data collection as shown in Fig. 1.
Country N Teaching Levels

Estonia 9 2 @ < 5 years, 3 @ 5-10 years Grades 2-12
Latvia 6 1 @ 11-15 years, 5 @ 20+ years Grades 4-12
Sweden 9 3 @ < 5, 2 @5-10 years, 2 @11-15 years, 2 @ 16-20 years Grades K-12
Finland 7 1 @ < 5, 2 @16-20 years, 4 @ 20+ years Grades 1-12
Lithuania 14 6 @ 5-10 years, 1 @ 16-20, 4 @ 20+ years Grades 1-12
Fig. 1. Survey II respondents
3.1 Data Analysis
The data collected in survey I on teacher’s associations between a range of terms

and conceptual categories, Computational Thinking, Information Technology
and so forth were summarized and plotted as a heat map to reveal the patterns
of association in the survey population.
Both surveys also collected qualitative data in the form of responses to a
number of free text questions. These data have been coded using a thematic
coding scheme using the TAMS software. The results presented in this paper
draw on a detailed analysis of the data collected in Survey II in response to
two questions which asked teachers to reflect on teaching challenges they had
experienced and teaching practices they adopted.
The thematic coding focused on capturing teacher attitudes and teaching con-
tent, and was conducted in accordance with standard inductive coding principles,
as described by Mayring [15]. Three iterations of classification were conducted
by the primary author to arrive at the clusters presented in Sect. 3.2.
3.2 Results and Discussion

Preparedness: The data collection includes a total of 504 responses of which
144 are complete and 360 are primarily demographic information about the
respondents. Lithuanian teachers dominate the sample, contributing approx-
imately two thirds of the responses, Swedish teachers comprise most of the
remaining third and Finnish teachers approximately 6% of the total sample.
The final question which asks respondents to associate terms and concepts with a
number of main focus areas, Computational Thinking, Computer Science, Infor-
mation Technology, Information and Communication Technology, Programming,
Mathematics, Algorithmic Thinking was answered by 144 respondents.
The majority of the respondents were directly involved in teaching comput-
ing and IT concepts in their schools. Teachers from nearly all year levels are
represented in the sample (see Fig. 2 for the distribution between year levels).
The sample also includes many very experienced teachers, the mean number
of years in teaching service in schools was 17 years among respondents, and
approximately 60% of the respondents were over 41 years of age (30% in the
age group 41–50 years). Most respondents felt that computing skills were impor-
tant ranking them as “vital” (40%) “very important” (40%) and “somewhat
important” (20%).
Fig. 2. Respondents teaching level
The curriculum in Sweden, Finland and Lithuania in 2016 emphasized the

teaching of CS content, especially programming in the last three years of school
education, typically years 10 to 12. The responses to Survey I reinforce this, most
of the teachers with active experience in teaching computing related content were
also heavily engaged in teaching at the upper secondary and middle school level.
While the largest volume of teaching is clearly in the years 10 to 12 of the school
system, there is also significant amounts of CS being taught in years 5 to 9.
Analysis: Many respondents felt that their school provided a somewhat, or

very, supportive environment for their efforts (78%). While this is encouraging
these teachers were already predominantly engaged in teaching technology, com-
puting mathematics or natural science subjects in their schools, and are also from
148 A. Pears et al.
countries where national curricula include CS as a formal subject. In the light of

the background of the respondents this result, while reassuring for Lithuanian
and Finnish school policy makers, is perhaps not particularly surprising.
A heat map which maps intensity of association in the response group of
terms and concepts to a range of higher level aggregate terms is presented in
Fig. 3.
Mobile Device Use

Scratch
Hour of Code
Binary
Coding
WordProcessing
ProbSolving
Algorithms
Recursion
Spreadsheets
Databases
Web Page (create)
Scripting
Animations
3D−modelling
Simulation Values
Computer Game (play) 110
Computer Game (create)
Graphics
ProgSewMachine 82.5
DrawGeoFigure
LEGO Mindstorms 55
Bee Bots
Robotics 27.5
Arduino
RaspberryPi 0
Internet
Moodle
PowerPoint(show)
PowerPoint(create)
OS
NetworkingConcepts
CompAdmin
TCP/IP
HTTP
Multi−core/Parallel
Variable
ProcFunc
Class
OOP
FuncProg
CT
CS
IT
ICT
Programming
Mathematics
Algoritms
NoAssoc
Fig. 3. Heat map of teachers’ concept associations
Teachers clearly had a clearer set of associations with Programming than with
many other terms and concepts listed in the table. Interestingly mobile devices Use
was quite strongly associated with IT and ICT, as was Internet. Bee Bots were not
clearly associated with any area, which may imply that the use of these devices to
reinforce relevant concepts is unclear to teachers. This warrants further investiga-
tion, since robotics was clearly associated with programming, algorithms and CT
or CS by many teachers in the sample. Problem solving and algorithms as were
strongly associated with programming, mathematics and algorithms which rein-
forces the conclusion that teachers consider these concepts as a knowledge cluster.
The only strong association with Word Processing was IT, which might reflect
the emphasis on tools such as Power Point and Word in the Swedish school IT
curriculum. Recursion was associated with programming and algorithms, but less
strongly with CS and mathematics, another area which may motivate additional
research.
Identifying Challenges: The sampling of teachers at schools in Estonia,

Latvia, Finland, Sweden and Lithuania resulted in 44 responses. The number
of responses from each country is shown in Fig. 1. The answers to question X are
indicative of a general consensus among the respondents that computing content
is important, and should be included in the school curriculum.
Our data reveals a central concern among the teachers sampled related to
student conceptual development and the relationship between the materials and
examples used to teach computing and other subject knowledge such as mathe-
matics and statistics. Historically speaking the traditional applications used to
teach computing concepts are dominated by examples drawn from Mathemat-
ics, one of the disciplines from which Computing emerged during the 1960’s and
1970’s [16]. The emergence of computing in schools requires new resources be
developed to meet the needs of contextualizing learning of computing for stu-
dents of ages from 5 to 18. This is expressed concisely by one of our respondents,
who comments as follows.
“Younger students didn’t find appealing making graphs in Excel because
they don’t know much about statistic in math yet. I must make graphs
about topics that are relatable [sic] to their everyday.”
In addition new classroom resources are needed to support teachers as they

introduce computational thinking content into subjects like languages, arts and
crafts, and technology.
An in depth thematic analysis of teachers motivations for teaching computing
content in their classes results in two dimensions motive and goal. In the moti-
vational dimension values identified are intrinsic, profession, societal. While in
the goal dimension values are associated with the pupil and focus on aspects of
cognitive development, enrichment, empowerment, future agency, and workplace
skills. Many teachers express an intrinsic motivation to engage with teaching
computing. The second greatest motivator is a sense of professionalism, this is
expected of them in their teaching role, and is important for the future of society.
However, many of the teachers also felt that they were compelled to engage
with the challenges of introducing computing competencies into their educa-
tional activities. Many were also clearly uncertain about their abilities. What
prompted feelings of insufficiency and failure among our sample of teachers?
Further analysis was conducted focusing on data collected in response the fol-
lowing question.
“Describe at least one teaching activity or lesson where you felt that you
failed in introducing aspects of computing in your teaching. What made
the experience feel like a failure? What did you learn from it?”
Here the majority of responses indicate that lack of appropriate examples and
teaching materials, as well as the need to teach outside one’s personal comfort
zone are the two areas of greatest difficulty. A typical statement in this regard
describes the challenges experienced by many of our participants.
150 A. Pears et al.
“I am not at all as in control of the situation as I usually am, since I am

not a particularly good programmer myself. The students manage to solve
most problems themselves, but sometimes it feels very frustrating not to
be able to help them when they get stuck.”
This highlights the need for in service training in programming for teachers who
engage in more advanced programming based classroom exercises.
4 Addressing Challenges with Action

The NordNICE project established a number of direct actions to support the
lack of support perceived by teachers. The very low levels of computing content
teaching in early school years reflect the current status of the curriculum, but also
highlight a potential lack of resources, since there is little incentive to produce
teaching materials for the current low volumes of pupils. The NordNICE project
has directly addressed this need.
The project repository collects teaching materials from the project school
partners helping to create a Nordic and Baltic repository of educational resources
which can support teachers at all school levels. These materials include class-
room exercises, resource books for teachers, and examples of good practice in a
range of disciplinary application areas from Art and Craft courses to Technology,
Mathematics and Physics.
To address the issue of accessible resources for younger learners the project
has developed translations of Bebras task cards for different age groups into
several European languages, including Finnish, Swedish, English and German.
While the NordNICE project conducted a series of in-service training activ-
ities for teachers, and organized network events where groups of teachers from
the participating countries made site visits to each other’s schools and met to
exchange ideas, and observe practices, much more focus on such initiatives is
needed. The teachers participating in our initiative were very positive to the
experience, and emphasized the importance of building up support networks and
training for themselves and their colleagues through similar national and inter-
national initiatives. Larger national programmes should be created to support
teacher’s professional development over the next five years as the new curriculum
is introduced.
5 Conclusion
The development of holistic curricula for computational thinking and comput-
ing concepts and techniques presents many opportunities and challenges. The
increasing dependence of our society and citizens on digital services, products
and systems makes the revision of curricula to include new skills which empower
future generations to manage a digital world vital to our future wellbeing.
This paper provides a Nordic and Baltic school teacher perspective on these
challenges and in particular how well governments, the teaching profession and
teacher training institutes are meeting these challenges. Our results indicate
that there is a competent cohort of in-service teachers who are very capable of
managing the transition to the new curricula. However, many of these teachers
have more than twenty years in the profession and are nearing retirement age.
This will leave a considerable competence gap to be filled in many countries (see
Fig. 1).
At the same time, many teachers experience a lack of support, in particu-
lar access to appropriate teaching materials which contextualize computing into
their subject area. The NordNICE project is a first step towards an integrated
approach that draws together expertise from Computing Education Research in
Higher Education, in-service teachers as mentors and coaches, and teacher train-
ing programmes to generate new approaches and provide access to innovative
materials.
While some progress has been made, much still needs to be done. Our work
suggests that it will be necessary to restructure teacher training programmes
in all subjects to provide appropriate computational thinking and computing
content knowledge to future teachers. Without such an investment the new cur-
ricula are at risk and pupils left without an appropriate high quality education.
Failure to attend carefully to these issues, especially failure to invest in teacher
training, places the future industrial competitiveness of the Nordic and Baltic
regions at risk.
Acknowledgements. The authors gratefully acknowledge the support of the Nordic

Research Council through the NordPLUS programme of transverse actions. In partic-
ular the funding of the two year network initiative, NordNICE, under project code
NPHZ-2015/10153.
A Appendix: Questionaires
The questionaires used to collect the data used in this paper are available on the
following link. Questionaires
References
1. Papert, S.: Mindstorms: Children, Computers, and Powerful Ideas. Basic Books
Inc., New York (1980)
2. Tedre, M., Denning, P.J.: The long quest for computational thinking. In: Koli
Calling, pp. 120–129 (2016)
3. Wing, J.M.: Computational thinking. Commun. ACM 49(3), 33–35 (2006)
4. Wing, J.M.: Computational thinking. In: VL/HCC, p. 3 (2011)
5. Benaya, T., Zur, E., Dagiene, V., Stupuriene, G.: Computer science high school
curriculum in israel and lithuania-comparison and teachers’ views. Baltic J. Mod.
Comput. 5(2), 164 (2017)
6. Dagiene, V., Stupuriene, G.: Bebras-a sustainable community building model for
the concept based learning of informatics and computational thinking. Inform.
Educ. 15(1), 25 (2016)
152 A. Pears et al.
7. Gander, W., Petit, A., Berry, G., Demo, B., Vahrenhold, J., McGettrick, A.,
Boyle, R., Mendelson, A., Stephenson, C., Ghezzi, C., et al.: Informatics educa-
tion: Europe cannot afford to miss the boat. ACM (2013). http://europe.acm.org/
iereport/ie.html
8. Hubwieser, P., Armoni, M., Giannakos, M.N., Mittermeir, R.T.: Perspectives and
visions of computer science education in primary and secondary (k-12) schools.
ACM Trans. Comput. Educ. (TOCE) 14(2), 7 (2014)
9. Rolandsson, L., Skogh, I.B.: Programming in school: Look back to move forward.
Trans. Comput. Educ. 14(2), 12:1–12:25 (2014)
10. Clear, T., Daniels, M., Cajander, Å., Parsjö, E., Lagerqvist, N., McDermott, R.:
A framework for writing personal learning agreements. In: IEEE/ASEE Frontiers
in Education Conference (2016)
11. Dagienė, V.: Teaching information technology and elements of informatics in lower
secondary schools: curricula, didactic provision and implementation. In: Mitter-
meir, R.T., Syslo, M.M. (eds.) ISSEP 2008. LNCS, vol. 5090, pp. 293–304. Springer,
Heidelberg (2008). https://doi.org/10.1007/978-3-540-69924-8 27
12. Liukas, L., Mykkanen., J.: Koodi 2016 - ensiapuaohjelmoinnin opetukseen perusk-
oulussa, June 2016. http://www.koodi2016.fi
13. Riis, U., ed.: IT i skolan mellan vision och praktik: En forskningsöversikt. Number
BEST. NR. 00:560 in ISBN 91-89313-89-5. Liber Distribution, Publikationstjänst,
162 89 Stockholm, Skolverket (2000)
14. Mannila, L., Dagiene, V., Demo, B., Grgurina, N., Mirolo, C., Rolandsson, L., Set-
tle, A.: Computational thinking in k-9 education. In: Proceedings of the Working
Group Reports of the 2014 on Innovation & #38; Technology in Computer Science
Education Conference. ITiCSE-WGR 2014, pp. 1–29. ACM, New York (2014)
15. Mayring, P.: Qualitative content analysis. Forum Qualitative Sozialforschung /
Forum: Qualitative Social Research 1(2) (2000)
16. Tedre, M.: The Science of Computing: Shaping a Discipline. Taylor & Francis,
Boca Raton (2014)
Computational Thinking/Programming,
Computational Abilities
Identifying Students’ Misconceptions on Basic
Algorithmic Concepts Through Flowchart
Analysis
Ebrahim Rahimi1(B) , Erik Barendsen2 , and Ineke Henze3

1
Radboud University, Nijmegen, The Netherlands
e.rahimi@cs.ru.nl
2
Radboud University and Open University, Nijmegen, The Netherlands
e.barendsen@cs.ru.nl
3
Delft University of Technology, Delft, The Netherlands
f.a.henze-rietveld@tudelft.nl
Abstract. In this paper, a flowchart-based approach to identifying sec-

ondary school students’ misconceptions (in a broad sense) on basic algo-
rithm concepts is introduced. This approach uses student-generated flow-
charts as the units of analysis and examines them against plan compo-
sition and construct-based programming problems to identify students’
misconceptions. In this study, 102 flowcharts, generated by 50 students
in two informatics classes in the Netherlands, were examined and var-
ious sorts of misconceptions were identified. The results suggest that,
given their abstract and language-independent nature, flowcharts can be
considered as an effective tool for revealing students’ difficulties in under-
standing algorithmic concepts. Our approach contrasts the more tradi-
tional use of program code to investigate students’ misconceptions. We
found several misconceptions mentioned in the literature, together with
two misconceptions which appear not to have been described before. Our
research contributes to the usage of flowcharts as a formative assessment
tool, directing informatics teachers’ instruction toward resolving these
misconceptions.
1 Introduction
There is an ongoing global debate among educational policy makers, researchers,

technology developers and practitioners around establishing computer science
(CS) and in particular computer programming as an integral part of school
curricula at different levels [1,6]. The advocates argue that due to the highly
technology-driven and information-dependent nature of working and living in the
21st century, learning fundamental computer science concepts and programming
has become a must for everyone [6]. They argue that CS represents an essential
and unprecedented sort of literacy people need to survive in the knowledge era
where thinking critically and computationally to solve complex, ill-defined, and
open-ended problems is becoming more and more important [21]. This sort of
https://doi.org/10.1007/978-3-319-71483-7_13
156 E. Rahimi et al.
literacy is meant to help people to not simply become technology consumers,

but also to serve as technology creators and knowledge developers [6,10].
Programming generally includes two main tasks: algorithmic problem solv-
ing and coding. The problem solving part includes analyzing the problem, for-
mulating a solution, and devising an algorithm for implementing the solution.
The coding part consists of implementing the devised algorithm using a pro-
gramming language, debugging and modification [15]. Algorithmic thinking thus
plays an essential role in problem solving and programming and accordingly has
a key position in computational thinking notions [21]. Despite the important role
of problem solving and algorithmic thinking in programming, in general, these
aspects seem to have received less attention than coding within CS in secondary
education [18].
Flowcharts are advocated as useful pictorial representations of algorithms,
the program logic or its flow of control. They have been an integral element
of programming since the introduction of computers in 1940s [15]. Flowcharts
have been used as simple and effective tools to comprehend, devise, modify,
visualize, debug, express and communicate algorithms and programs [12]. As
remarked by [15], a flowchart represents “a high level definition of the solution
to be implemented on a machine” (p. 1). Given their abstract and language-
independent nature, flowcharts have been suggested as effective tools for the
novice programmers to enhance their problem solving and algorithmic thinking
competencies, and to promote a “thinking first” approach to programming [18].
As held by [2], flowcharting helps “distinguish between the procedure a computer
program is written to express and the syntactical details of the language in which
the program is written” (p. 53).
The purpose of this study is to investigate the usability of flowcharts as a
tool for identifying and highlighting students’ misconceptions on basic algorith-
mic concepts. Following [20], we will use the term misconception as an overall
term covering students’ lack of understanding, problems, mistakes, bugs, and
difficulties with basic algorithm concepts. Flowcharts can be seen as a way to
get into the students’ minds and provide a better picture of their understand-
ing of basic computer science concepts. By doing so, flowcharts ultimately serve
to improve the quality of computer science education through a better under-
standing of what goes wrong [20]. In [18] an analytical approach to identifying
programming misconceptions using flowcharts is introduced. This approach uses
two categories of plan composition and construct-based problems introduced
by [20] to detect students’ misconceptions. It was used to detect programming
misconceptions exhibited by 11 high school students in the Netherlands.
Our study adopted the approach introduced in [18] and modified it to
detect students’ misconceptions on each of basic algorithmic concepts including
sequence, condition, iteration, and sub-algorithm. To this end, 102 flowcharts
generated by 50 students in two informatics classes in the context of upper sec-
ondary education in the Netherlands were examined.
This study is a part of a bigger research project with the main objective of
encouraging and facilitating the learning of computer science concepts through
Identifying Students’ Misconceptions on Basic Algorithmic Concepts 157
conducting authentic design assignments (see [9]). This study plays a multifold
role in this research project: (i) highlighting the importance of student-generated
flowcharts as intermediate design products, (ii) examining the usability of flow-
charts as an effective formative assessment tool, (iii) helping the teachers and
researchers in identifying students’ misconceptions on basic algorithm concepts,
and (iv) directing teachers instruction toward resolving these misconceptions.
This way, applying flowcharts as a formative assessment tool might help to
further develop the teachers’ PCK (Pedagogical Content Knowledge) [16] on
algorithms.
2 Programming Misconceptions
There are worldwide observations that students, regardless of institution or coun-
try, show generally poor performance in the introductory programming courses
and learning programming is a difficult task for novices [3,17]. A reason for this
difficulty stems from the fact that programming puts a high cognitive and knowl-
edge demand on novices including knowledge on a specific programming language
and knowledge and understanding of basic programming concepts and constructs
such as variables, loops, conditions, abstraction, and procedures. Any misconcep-
tion on these constructs might result in programming difficulties [17,18]. There-
fore, realizing and resolving these misconceptions seems useful in diminishing
the novices difficulties in programming.
Students’ misconceptions on basic programming constructs and features have
been vastly researched (see [4,8,11,14,17,19,20]). Four common examples of the
identified programming misconceptions include considering classes as containers
for objects, reverse assignment, interpreting assignment statements as mathe-
matical equations, and boundary problem (i.e., choosing inappropriate bound-
ary points) [17,20]. Seppälä et al. in [13] categorized students’ errors on algo-
rithm exercises into systematic and careless errors (‘slips’). According to [13],
slips result from randomly trying out algorithm exercises by students, whereas
a systematic error is “a symptom of a misconception that could be corrected if
recognized” (p. 244). According to [20], students’ misconceptions on program-
ming fall in two categories: plan composition and construct-based problems. Plan
composition problems refer to the encountered misconceptions in composing a
solution by putting the pieces of plans together. These misconceptions are con-
cerned with the solution formulation, algorithm development, and planning the
semantic and logic of the program [3]. On the other hand, construct-based prob-
lems concern with the misconceptions on language constructs and the syntax of
the program. Tables 1 and 2 summarize plan composition and construct-based
problems introduced by [18,20].
However, the majority of these misconceptions have been researched and
recognized in the context of program’s code in specific programming languages.
There are very few studies examining the programming misconceptions at a more
abstract level such as algorithms and flowcharts (see [18]). Arguably, the sorts
of misconceptions exhibited by students in a programming language depend, to
Table 1. Plan composition problems, adopted from [20]
Misconception Description
Summarization The complex combinations of plans are summarized in
problem terms of some primary functions and secondary functions
have been overlooked
Optimization Novices aim to optimize their plans but do not adequately
problem check if the optimization can really be carried out
Previous-experience Novices constantly develop and tailor plans on the basis of
(or pollution) previous experience. This error is introduced when
problem inappropriate aspects of previous plans pollute a related
plan that is being used in a new situation
Specialization Inappropriate and incorrect customization of an abstract
problem plan developed for other situations
Natural-language Errors happen during the process of mapping from natural
problem language to a programming language.
Boundary problem Novices difficulties for deciding on appropriate boundary
points in specializing a plan
Unexpected case The program is not working correctly for all cases (e.g.
problem uncommon, unlikely, or boundary cases)
Interpretation Not considering the implicit specifications of plans or
problem misinterpreting them
Cognitive load Omitting and overlooking small but important parts of the
problem plan or plan interactions
Table 2. Construct-based problems, adopted from [18, 20]
Misconception Description
Human interpretation problem Novices assumption that computers are able to
interpret problems as people do
Assignment misconception Inverted assignment: the positions of the giver
and receiver variables at the right and left side of
the assignment operator are misplaced
Condition misconception Misconceptions about how the condition construct
(or control structure) works
Boolean statements Misconceptions on how boolean statements behave
Loops Misconception on how a loop control variable
works
Method-related misconceptions Misconceptions on method calling
Control flow Misconceptions such as incorrect use of print and
return statements, omitting arrows in a flowchart,
or omitting start or stop steps
some extent, on the specifications of that programming language and might be

shaped or filtered by its structures and specifications. In contrast, flowcharts
provide a more abstract and language-independent way to get into the minds of
students and seem useful in gaining a firsthand and more comprehensive picture
of what goes wrong in their minds.
3 Study Setting
The participants in this study were 50 students of age 15 or 16 from two schools
in upper secondary education in the Netherlands. All of the participants were
at the university preparatory education level (VWO in Dutch) [7]. In the afore-
mentioned surrounding research project (see the introduction section), the par-
ticipants were provided with the background and design documents. The back-
ground document covered various aspects of algorithms including the defini-
tion, basic concepts (i.e. sequence, condition, iteration, and sub-algorithms),
trace table, and flowcharts. Each of these aspects was explained using several
examples and questions with a different level of difficulty. The design document
included seven authentic design assignments in the context of text analysis. In
each assignment, students were asked to follow a step-wised approach to learning
algorithms and programming. These steps include analyzing the design problem,
formulating a conceptual solution, devising an algorithm, developing a flowchart
to implement the algorithm, testing the flowchart using trace tables, and con-
verting their flowcharts into a program using a programming language (either
PHP or Python, as the students had learned one of these languages in the previ-
ous years). Students could use the background document whenever necessary for
explanation and advice on conducting these steps. To embed formative assess-
ment within this educational intervention, students were asked to answer three
exit questions during their design endeavours on a weekly basis. These questions
were derived from the background document and were meant to assess students’
understanding of basic algorithm concepts underpinning the design assignments.
In each question, the students were asked to draw a flowchart for solving the given
problem. To foster collaboration and team working, the students were grouped
into teams of 2–3 students and each team developed a flowchart for each of the
questions. The flowcharts were written on paper or developed digitally using a
specific tool called Draw.io and were handed over manually or via the school’s
learning management system. The teachers then could use these flowcharts to
detect students’ misconceptions and provide them with appropriate feedback.
Table 3 presents these questions and their associated algorithm concepts.
The following research question directed the data collection and analysis
processes: What misconceptions can be seen in students’ flowcharts?
The flowcharts developed by the participating students in response to the
questions in Table 3 were used as the data source to answer this research ques-
tion. A qualitative deductive content analysis approach was followed to analyze
the flowcharts. Content analysis is a method used for analyzing written, verbal
or visual communication messages and documents [5]. Following the process of
Table 3. Questions asked to investigate students’ understanding of basic algorithmic

concepts
Question Concepts
Q1 : Write an algorithm that receives three Sequence, condition
numbers (a, b, c) and determines and reports the
maximum and minimum numbers
Q2 : (i) Write two sub-algorithms findmax (a,b) Sub-algorithms, problem
and findmin (a,b) which receive two numbers a, b composition and
and determine and return the maximum and decomposition, condition
minimum numbers, respectively (ii) Using these
sub-algorithms write an algorithm that receives 10
numbers (i.e., a0 , a1 , ... , a9 ) and determines and
reports the maximum and minimum numbers
Q3 : Write an algorithm that receives a text value Loop, condition
(i.e. t) as input and reverses it (i.e. t = “book”,
reverse of t = “koob”)
deductive content analysis introduced by [5], three steps of preparation, organi-

zation, and reporting were taken. In the preparation step, all of the flowcharts
(102 in total) as the unit of analysis were uploaded into Atlas.ti software. In
the organizing phase, first the plan composition and construct-based problems
presented in Tables 1 and 2 were chosen as the initial categories for coding the
observed misconceptions in the flowcharts. Then following an iterative process of
coding and revising, the misconceptions on basic algorithm concepts observed in
the flowcharts were coded according to these categories. The coding process was
flexible to allow emerging new codes which did not exist in these categories. Pos-
sible alternative interpretations of the observed misconceptions were discussed
within the research team until a consensus was reached. Finally, the identified
misconceptions on each of basic algorithm concepts were processed and reported
as can be seen in the next section.
4 Results
Tables 4 and 5 present the identified plan composition and construct-based prob-
lems in the flowcharts, respectively.
• Plan composition problems
(i) Condition misconceptions: Three plan composition misconceptions for the
condition concept were identified. Not considering equal inputs in plans is an
unexpected case problem where the equal numbers are excluded from a solution,
specifically for answering questions 1 and 2 in Table 3. This was one of the most
frequently observed misconceptions in the flowcharts. Interestingly, some of the
students reacted to passing equal numbers to their flowcharts as a fatal error (see
Fig. 1(a)). Incorrect adaptation is the next condition misconception and repre-
sents a previous experience problem. Figure 1 (b) illustrates an example of this
Table 4. The plan composition problems observed in the flowcharts
Category Misconceptions
Condition misconceptions Not considering equal inputs in plans
(unexpected case problem): the situations
where the equal numbers are excluded from
the comparison operations
Incorrect adaptation (previous experience
problem): refers to incorrect applying of a
previous condition-related plan in another
situation
Misinterpreting the condition concept
(Interpretation problem)
Loop misconceptions Incorrect loop construction
Sub-algorithm misconceptions Composition problem: refers to students
problems and mistakes with composing an
algorithm by putting together
sub-algorithms
Translation Mapping problem: incorrect mapping from
natural language or programming languages
to flowcharts
misconception where an incorrect adaptation of the plan used for determining

the maximum number (i.e. max ) between a, b is used for calculating the mini-
mum number (i.e. min). Misinterpreting the logic of condition concept is a sort
of interpretation problem with regard to the condition concept, as depicted by
Fig. 1 (c).
Fig. 1. Examples of identified plan composition problems related to the condition

concept
(ii) Iteration misconceptions: Incorrect loop construction is the encountered

misconception related to the implementation of the iteration concept. Figure 2
depicts an example of this misconception.
Fig. 2. An example of incorrect loop implementations
(iii) Sub-algorithm misconceptions: The composition problem refers to students

problems, difficulties, and mistakes with composing an algorithm using its sub-
algorithms. The results showed that while many of the students devised the
asked sub-algorithms (i.e. findmax and findmin) correctly, many of them were
not able to use the developed sub-algorithms to compose the main algorithm.
(iv) Translation misconceptions: The mapping problem or incorrect translation
from formal language or programming languages to flowcharts was another plan
composition problem observed in the flowcharts. Figure 3 illustrates an example
of this misconception. Surprisingly, as shown by this example, although some of
the students already know how to program, but they cannot map their program
into a flowchart.
Fig. 3. An example of difficulty in mapping a plan from programming languages to a

flowcharts
• Construct-based Misconceptions
(i) Condition misconceptions: Missing the false part represents a misconception
where only the true output of a condition statement is addressed. Figure 4 (a)
presents an example of this misconception. The 3-output condition is another
misconception where 3 possible outputs are considered for a conditional state-
ment, as shown in Fig. 4 (b).
(ii) Sequence misconceptions: The sequence misconceptions were among the fre-
quent problems observed in the flowcharts. These misconceptions reflect stu-
dents’ misunderstanding about the flowchart’s flow control and sequential exe-
cution of its steps. The parallel execution misconception reflects a perception
Fig. 4. Example of construct-based misconceptions for the condition concept
held by some of the students that the flowchart’s steps can run in parallel to
each other, as shown by Fig. 5 (a). Another observed misconception with the
sequence concept is called the dense step which refers to a flowchart’s step that
executes more than one operations. Figure 5 (b) shows two examples of this
misconception. It seems that in these examples the students assumed that the
flowcharts interpret these steps like what people do. The last sequence miscon-
ception concerns with students’ issues with Input/output and start/stop steps.
These issues refer to situations where students forgot to get input, report or
return the final results, forgot to add start and stop steps in their flowcharts, or
used return instead of report.
Fig. 5. Examples of construct-based misconceptions for the sequence concept
(iii) Loop misconceptions: Figure 6 illustrates an example of the identified

construct-based loop misconception. This misconception is concerned with the
simultaneous initialization and check of the loop’s control variable. As can be
realized from this example, this misconception might lead to either infinite or
zero-repeating loops, depending on the value of len(word1) - 1.
(iv) Assignment misconceptions: Two construct-based misconceptions related to
the assignment operation were discerned in the flowcharts. The most occurring
error in the students flowcharts was the inverted assignment where the positions
of the source and destination variables in the right and left of the assignment
operator were misplaced. Figure 7 (a) shows an example of this misconception.

In this example the students were asked to find the maximum and minimum
numbers among variables a, b, c and store them in variables max and min,
respectively. Another type of the observed assignment issues is the missing value
misconception where a value is processes but not stored in a variable. Figure 7
(b) shows an example of this misconception.
Table 5. The construct-based problems observed in the flowcharts
Category Misconceptions
Condition misconceptions Missing the False part: conditional statements
where the false part (or the else part) is missing
3-output condition: conditional statements with 3
outputs
Sequence misconceptions Dense step (Human interpreter problem):
situations where more than one operations are
processed in one step
Parallel execution of steps (flow control)
Issues with Input/output and start/stop steps
(flow control)
Loop misconceptions Simultaneous initializing and checking of the loop
control variable
Assignment misconceptions Inverted assignment: situations where the
positions of the source and destination statements
of the assignment operator are misplaced
3-output condition: conditional statements with 3
outputs
Sub-algorithm misconceptions Incorrect use of sub-algorithms: for instance using
a sub-algorithm at the left side of an assignment
operator
Issues of calling/returning from sub-algorithms
Flowchart presentation Using incorrect shapes for the flowchart’s
constructs
Fig. 6. An example of the loop misconception

Fig. 7. Two examples of the assignment misconceptions
(v) Sub-algorithm misconceptions: The Incorrect use of sub-algorithms exhibits

an misunderstanding and wrong interpretation of the usage of sub-algorithms.
For instance, it was observed that some of the students interpreted a sub-
algorithm as a variable or container and used it at the left side of an assign-
ment operator. Issues of calling/returning from sub-algorithms represent another
sub-algorithm misconception. Examples of this misconception include: not pass-
ing input parameters to sub-algorithms and not returning results from sub-
algorithms.
(vi) Flowchart presentation: This misconception refers to situations where incor-
rect shapes were used by the students to present various constructs of their
flowcharts. For example, several students used rectangle instead of diamond for
presenting conditional statements.
5 Conclusions and Discussion

In this study, a flowchart-based analysis approach, adapted from [18,20], was
introduced and followed to identify the misconceptions of basic algorithm con-
cepts exhibited by 50 students in upper secondary education in the Netherlands.
Various types of plan composition and construct-based misconceptions on basic
algorithm concepts have been identified in the analyzed flowcharts. The identi-
fied plan composition misconceptions are: not including equal numbers in plans
(an unexpected case misconception), incorrect adaptation (a previous experience
misconception), misinterpreting the condition concept (an interpretation mis-
conception), incorrect loop construction, composition problem (i.e. devising an
algorithm using sub-algorithms), and mapping problem. The discerned construct-
based misconceptions include missing the false part of conditional statements, 3-
output condition, parallel execution of steps (a flow control misconception), dense
step (a human interpreter misconception), missing input/output and start/stop
steps, incorrect initialization and checking of the loop counter, inverted assign-
ment, missing values, incorrect use of sub-algorithms, issues of calling/returning
from sub-algorithms, and using incorrect shapes for the flowchart’s statements.
Computer programming consists of two interconnected contexts of high level,
abstracted and language-independent algorithmic thinking, and detailed, step-
wised, and language-dependent coding. Our findings suggest that some of the
programming misconceptions at the abstract level might be filtered by specifi-
cations of programming languages and accordingly not be visible for language-
based misconceptions detecting approaches. On the basis of our findings, it can
be concluded the flowcharts, by inheriting and bridging between the essence and
key characteristics of algorithmic thinking and coding contexts, are useful means
to detect misconceptions exhibited by students in the solution formulation and
algorithmic thinking phases.
As observed in the results section, the students exhibited various sort of
plan composition and construct-based problems on basic algorithm concepts in
their flowcharts. The inverted assignment was the most occurring misconcep-
tion encountered by the students. The same finding has been reported by other
researchers (see for example [8,17]). Furthermore, the students showed several
misconceptions related to the sub-algorithm concept including the composition
of an algorithm using sub-algorithms. Arguably, a reason for this misconcep-
tion stems from this fact that the majority of the students did not have any
previous experience of working with abstract constructs such as sub-algorithms.
Surprisingly, it has been observed that some of the students first wrote computer
programs and then tried to convert their programs to flowcharts.
Given the programming misconceptions presented in Tables 1 and 2 as the
analytical framework, the results suggest that there are many overlaps between
misconceptions encountered by students in the algorithmic and coding parts
of the programming process. For example, human interpretation, previous plan
experience, unexpected case, and inverted assignment misconceptions happen
similarly in both parts [18]. However, there are two exceptions with regard to
the parallel execution and dense step misconceptions. These misconceptions rep-
resent some of the students’ assumption that flowchart’s steps can run parallel
to each other. To the best of our knowledge, these are newly identified mis-
conceptions and have not been reported in other studies. A possible reason for
this neglect might be due to this fact that the majority of research on program-
ming misconceptions has been conducted in a specific programming language
and the syntactical structure and specifications of the programming language
filter emerging this sort of abstract misconceptions.
The use of flowcharts generated by groups of students as data, instead of
individually generated flowcharts, can be seen as a limitation of the study. One
might claim that group dynamics might amplify or filter misconceptions aris-
ing at an individual level. A second limitation stems from conducting the study
with students who had previous experience of programming. Repeating the same
study by students with less programming experience might result in detection of
other kinds of misconceptions. Other possibilities for follow-up research include
using the identified misconceptions to develop CS teaching materials for the

algorithmic thinking and examining their effectiveness in resolving these mis-
conceptions, and scrutinizing possible links between students’ exhibited miscon-
ceptions in algorithmic thinking and coding phases of programming. As to the
bigger research project on design-based education, our results suggest that using
flowcharts to express intermediate design products could provide a more authen-
tic way to incorporate formative assessment of fundamental concepts during
students’ project work.
References
1. Barendsen, E., Grgurina, N., Tolboom, J.: A new informatics curriculum for sec-
ondary education in the Netherlands. In: Brodnik, A., Tort, F. (eds.) ISSEP 2016.
LNCS, vol. 9973, pp. 105–117. Springer, Cham (2016). https://doi.org/10.1007/
978-3-319-46747-4 9
2. Bohl, M.: Flowcharting Techniques. Science Research Associates, Chicago (1971)
3. Chetty, J., van der Westhuizen, D.: Towards a pedagogical design for teaching
novice programmers: design-based research as an empirical determinant for success.
In: Proceedings of the 15th Koli Calling Conference on Computing Education
Research, pp. 5–12. ACM (2015)
4. Clancy, M.: Misconceptions and attitudes that interfere with learning to program.
Comput. Sci. Educ. Res., pp. 85–100 (2004)
5. Elo, S., Kyngäs, H.: The qualitative content analysis process. J. Adv. Nurs. 62(1),
107–115 (2008)
6. Mannila, L., Dagiene, V., Demo, B., Grgurina, N., Mirolo, C., Rolandsson, L.,
Settle, A.: Computational thinking in k-9 education. In: Proceedings of the Work-
ing Group Reports of the 2014 on Innovation & Technology in Computer Science
Education Conference, pp. 1–29. ACM (2014)
7. Nuffic: The Dutch education system described (2015). https://www.nuffic.nl/en/
publications/find-a-publication/education-system-the-netherlands.pdf. Accessed
Sept 2017
8. Putnam, R.T., Sleeman, D., Baxter, J.A., Kuspa, L.K.: A summary of miscon-
ceptions of high school basic programmers. J. Educ. Comput. Res. 2(4), 459–472
(1986)
9. Rahimi, E., Barendsen, E., Henze, I.: Typifying informatics teachers’ PCK of
designing digital artefacts in dutch upper secondary education. In: Brodnik, A.,
Tort, F. (eds.) ISSEP 2016. LNCS, vol. 9973, pp. 65–77. Springer, Cham (2016).
https://doi.org/10.1007/978-3-319-46747-4 6
10. Rahimi, E., van den Berg, J., Veen, W.: Investigating teachers’ perception about
the educational benefits of Web 2.0 personal learning environments. eLearning Pap.
(35) (2013)
11. Robins, A., Rountree, J., Rountree, N.: Learning and teaching programming: a
review and discussion. Comput. Sci. Educ. 13(2), 137–172 (2003)
12. Scanlan, D.A.: Structured flowcharts outperform pseudocode: an experimental
comparison. IEEE Softw. 6(5), 28–36 (1989)
13. Seppälä, O., Malmi, L., Korhonen, A.: Observations on student misconceptions–a
case study of the build-heap algorithm. Comput. Sci. Educ. 16(3), 241–255 (2006)
14. Sheard, J., Simon, S., Hamilton, M., Lönnberg, J.: Analysis of research into the
teaching and learning of programming. In: Proceedings of the Fifth International
Workshop on Computing Education Research Workshop, pp. 93–104. ACM (2009)
15. Shneiderman, B., Mayer, R., McKay, D., Heller, P.: Experimental investigations of
the utility of detailed flowcharts in programming. Commun. ACM 20(6), 373–381
(1977)
16. Shulman, L.S.: Those who understand: knowledge growth in teaching. Educ.
Researcher 15(2), 4–14 (1986)
17. Sirkiä, T., Sorva, J.: Exploring programming misconceptions: an analysis of student
mistakes in visual program simulation exercises. In: Proceedings of the 12th Koli
Calling International Conference on Computing Education Research, pp. 19–28.
ACM (2012)
18. Smetsers-Weeda, R.: Think... then act() flowcharts as a tool for novice program-
mers to enhance their problem solving skills. Master’s thesis, TU Delft, The Nether-
lands (2016)
19. Sorva, J.: Notional machines and introductory programming education. ACM
Trans. Comput. Educ. 13(2), 8 (2013)
20. Spohrer, J.C., Soloway, E.: Novice mistakes: are the folk wisdoms correct? Com-
mun. ACM 29(7), 624–632 (1986)
A Preliminary Investigation
on Computational Abilities in Secondary School
Ugo Solitro1(B) , Margherita Pasini2 , Debora De Gradi1 ,

and Margherita Brondino2
1
Department of Computer Science, Università Degli Studi di Verona, Verona, Italy
ugo.solitro@univr.it
2
Department of Human Sciences, Università degli Studi di Verona, Verona, Italy
{margherita.pasini,margherita.brondino}@univr.it
Abstract. For many years, several countries have been committed to

the introduction of computer science and, more recently, computational
thinking in primary and secondary schools. Nevertheless, students of the
first year of university often encounter difficulties in dealing with intro-
ductory courses in computer science and, in particular, programming.
One of the main issues is related to the difficulty of thinking about algo-
rithmic solutions, a skill that should be acquired from secondary school.
A good knowledge of difficulties and the discovery of efficient strategies
to overcome them are vital tasks. In this work, we discuss a first step in
this direction. We have prepared a brief test focused on the algorithmic
abilities and other processes potentially involved with it. The idea is to
build a useful tool that can give assistance to the introduction of compu-
tational thinking skills in secondary school education. The test has been
proposed to a small sample of secondary school students, aged 11 to 16
and its outcomes analyzed.
Keywords: Computational thinking · Algorithmic thinking · Problem

solving · Secondary school · Computing education
1 Introduction
When we observe the performance of a first year introductory course in program-
ming in a non vocational curriculum, we can be really disappointed: difficulties in
learning, low performances, high drop-out. One of the most relevant problems is
seemingly originated from too little familiarity with algorithm and effective pro-
cedures in the previous school experience. Many institutions in the past decades
have proposed and developed experimental projects for the introduction of infor-
matics and digital technology in School. This subject is also widely discussed in
the community (see for instance [8,18,19]). More recently, informatics and, in
particular, computational thinking have been incorporated into school curricula
starting from primary schools. In spite of this, the final outcomes are often poor.
Schools often do not have access to resources, notably people with adequate
experience but also time, computer facilities.
https://doi.org/10.1007/978-3-319-71483-7_14
170 U. Solitro et al.
Italy is engaged in a long process of reforming the School. This process

involves in particular the introduction of informatics in all degrees, with an
emphasis on digital competences instead of wide and cross-disciplinary compu-
tational skills. An account of this can be found in [8]. In 2015, more recent reform
(also known as “La Buona Scuola”) has indicated Computational Thinking as
one of priority objectives of the Italian school.
This work is part of a wider project aimed to study learning skills and
difficulties encountered by students - mainly secondary school and university
students - related to computational and algorithmic skills. This project also
involves university students enrolled at the first degree in Applied Mathematics,
and is oriented to drop out reduction. The project includes a deep examina-
tion of students’ learning difficulties in programming. As a part of the activities,
we also changed some teaching methodologies, analyzed motivation, evaluate
performance and examined their background [11,12,16,17]. In particular, we
discovered that the greater part of students lack of an adequate introduction
to computational skills1 . The project intends also to explore the modalities of
introduction to computational thinking, actual problem solving methodologies
and algorithmic skills, in middle and high schools. This study aims to present
the results of the assessment of a small set of preliminary skills in secondary
school students (grades 6 to 11).
This study presents the findings of the assessment of a small set of basic
skills with the aim of developing instruments that could help the introductions
of computational skills in the education and predicting difficulties.
2 Computational and Algorithmic Thinking

Computational Thinking (CT) is a permeating concept in the recent dis-
cussions about Computer Science Education. Papert in [10] suggests “to use
computational thinking to forge ideas that are at least as explicative as the
Euclid- like constructions (and hopefully more so) but more accessible and more
powerful.” This expression is more often related to the paper by Wing [20]
which describes CT as a “universally applicable attitude and skill set” that
includes solving problems, designing systems and understanding human behav-
iour using the fundamental concepts of computer science. On the other hand,
Knuth [6] describes Algorithmic Thinking using a set of characterizing cate-
gories: representation of reality, reduction to simpler problems, abstract rea-
soning, information structures, algorithms. These two abilities are, in fact,
strongly related and, in particular, CT requires the ability to identify appropriate
algorithms.
In a recent review [14], Plerou shows the classification of algorithmic think-
ing difficulties and mathematical learning difficulties, grouping them into 6
categories: visual perception, spatial and geometrical perception, time percep-
tion, calculation, calculus perception, algorithmic thinking. Algorithmic thinking
1
This datum is available by consulting both entrance test results and students’
answers to a survey proposed at the beginning of the course.
A Preliminary Investigation on Computational Abilities in Secondary School 171
deficits concern perception, design and description of an algorithm, best solution

evaluation and the ability to separate a problem into smaller sub-problems.
In the last years, several studies investigated the skills used in programming.
These studies used standardized tests to evaluate the required skills and students’
results in programming courses or programming tasks to check for hypothesized
verify the correlations.
Erdogan et al. [5] tested psychological predictors of 48 high school students.
The predictors tested are creativity, problem solving, computer attitude and
general skills, which include analytical thinking, abstract thinking and spatial
perception. The results show a significant correlation between programming and
mathematics grades, as well as between programming grades and general skills.
Milić [9] investigated on the IQ level, computer attitude and gender. This
work studies the correlation with two types of programming tasks, a low level
and a high level knowledge. The low level performance, which requires to per-
form a program, is correlated with IQ and computer attitude. The high level
performance, which requires the ability to create a program, is correlated only
with the low level performance and not with any of the other predictors tested.
Korkmaz’s in [7] shows that critical thinking and logico-mathematical intel-
ligence are both correlated with algorithmic skills, intended as problem analysis,
solution design and mathematical transfer.
Ambrosio et al. [2] tested general intelligence, spatial reasoning, mathemat-
ical reasoning and attention to details. The results suggest that general intelli-
gence and spatial reasoning are factors affecting the learning of programming,
as these two factors are related to the final grades of the programming course.
3 Method
3.1 Participants
The data are collected among 261 students, of which 139 middle (i.e. lower
secondary) school students (from 6 to 8 grades) and 122 high school students
(form 9 to 11 grades).
Middle school students were from a residential district of the city. The school
teaching plan prescribes 6 h of mathematics per week and no independent infor-
matics teaching. The “digital” background is heterogeneous, with generally low
CT skills. The sample is composed by 75 female students and 64 male students,
distributed among the classes as shown in Table 1.
Table 1. Gender distribution among the middle school classes.
Class 1A 1B 1C 2A 2B 3A 3B
M 13 10 9 4 6 11 11
F 8 11 9 9 15 12 11
High school students were from a technical institute (Electronics, Informatics

and Logistics) located in the city. The first two years are organized essentially in
the same way for all the three curricula and the specialization starts in the third
year. In Table 2 is synthetically described the group of participants with respect
to sex and specialization. High school students generally show a heterogeneous
“digital” that is dependent on middle schools, previous experiences and personal
interests. In the first two years, they have to attend specific introductory courses
in ICT. From the third year, informatics is considered more seriously as part of
a professional training.
Table 2. High school sample data.
Class Specialization # M F Courses

1 — 41 36 5 Mathematics (4)
Computer technologies (1 + 2)
2 — 35 35 0 Mathematics (4)
Applied science and technologies (3)
3 Informatics 24 24 0 Mathematics (3)
Complements of mathematics (1)
Telecommunication systems and technology (2 + 1)
Systems and networks (2+2)
Computer science (3 + 3)
Telecommunication (1 + 2)
Logistics 22 20 2 Mathematics (3)
Complements of mathematics (1)
Automation and electronic (1 + 2)
Logistics (3 + 2)
In parenthesis the hours per week dedicated to the course:
(classroom) or (classroom + laboratory).
#: number of students tested.
M: number of male students.
F: number of female students.
3.2 Procedure
The test consists of two different parts:
1. the first part is about the general abilities,
2. the second part regards computational and algorithmic skills.
Both parts were presented during normal class time, so the students were not
required to dedicate extra time to the test. The activity was set following the
need of the school and the availability of teachers, and hence not necessarily
during mathematics or computer science classes. The test required about 50 min
with an upper limit of a 1 h, depending on the school schedule and needs.
The first part of the test was performed in about 35–40 min. The students
received all the questions and an answer sheet on which they had to report their
answers and an identification number assigned randomly.
After the completion of the first part (all the answer sheets have been col-
lected), the second part, consisting in a single sheet of paper, was hand out to
students. In this case, the answers were marked directly on the paper, as well as
the identification number. The second part of the test was performed in about
15–20 min.
3.3 Instruments
Bebras international contests have come out as a successful way to introduce

computational thinking activities in School and Bebras [1] activities has been
proposed for the introduction of informatics in the curriculum [3,4]. Hence the
items in our test are partly based on Bebras tasks and partly on more traditional
tests.
The First part of the test includes 18 items. The aim is to verify the students
general abilities in 8 tasks that could be related to computational thinking.
The items involve non verbal aspects, pattern recognition, spatial orientation,
temporal order, procedure application, best solution identification, reasoning by
analogies and problem comprehension. The items used are based on the Bebras
tasks, the problem solving tasks and Raven’s Progressive Matrices (RPM) [15].
RPM is a collection of widely-used standardized intelligence tests consisting of
analogy problems in which a matrix of geometric figures is presented with one
entry missing, and the correct missing entry must be selected from a set of answer
choices.
The Second part acts as a criterion, useful to test the validity of the first part as
a set of preliminary skills connected with programming skills. This second part
is, in some way, estimates students’ programming skills. It is organized in two
“specific tasks”:
– a coding task, again inspired by Bebras activities, where the student has to
describe a simple procedure in a formalized language;
– an algorithmic task, based on the activities described in [13], where the stu-
dent has find a solution to a problem in an effective way.
Since the algorithmic task is more challenging, it has been organized in sub-
tasks guiding the student toward the solution.
4 Results
In this section we identify the tasks with an acronym that is connected with the
specific goal:
Acronym Description
R Non verbal (Raven matrix)
S Pattern recognition
O Spatial orientation
ORD Sequential order
P Procedure application
SO Best solution identification
A Reasoning by analogy
CT Problem comprehension
AL Problem solving
COM Coding
4.1 Correlations Between General Abilities and the Specific Tasks

To verify the effects of general abilities on the specific tasks, we study the correla-
tion between scores in each general task and scores in the final tasks, algorithmic
and coding.
The correlation matrix (Table 3) with Pearson’s coefficients shows significant
correlations between spatial orientation and coding (r = .292, p < .001), and
temporal order and coding (r = .299, p < .001). For the algorithmic ability, sig-
nificant correlations are shown with: spatial orientation (r = .355, p < .001), tem-
poral order (r = .342, p < .001) and best solution recognition (r = .355, p < .001).
Table 3. Pearson’s correlation matrix with significance level.
COM AL
R 0.15* 0.11
S 0.14* 0.24***
O 0.29*** 0.36***
ORD 0.30*** 0.34***
P 0.20** 0.21***
SO 0.20** 0.36 ***
A 0.12 0.21***
CT 0.07 0.19**
COM 0.20**
∗
p < .05, ∗∗ p < .01, ∗∗∗ p < .001
Then we tested two regression models to see which predictors better explain
coding task and algorithmic ability. Spatial orientation (β = .22, p < .001) and
temporal order (β = .21, p < .001) seemed to be the stronger predictors of cod-
ing task explaining 12% of variance. For algorithmic ability, spatial orientation
(β = .20, < .001), best solution recognition (β = .21, < .001), temporal order
(β = .20, < .001) and reasoning by analogies (β = .13, p = .024) explained 23%
of variance.
4.2 Specific Abilities Interaction

Since for the coding task we have only three possible scores, that are just 0,
1
2 or 1, the results obtained in the previous section about this task are not
really reliable. We are therefore interested in running a one way ANOVA using
as a fixed factor the coding score and as a dependent variable the algorithmic
score, this analysis makes sense since we have mentioned that the computational
ability includes the algorithmic ability. The ANOVA confirms our inclusion as
it shows a significant main effect of the computational coding performance on
the algorithmic score (F (2, 258) = 5.25, p < .01), in particular the post hoc
tests show significant difference (pbonf < .01) is found between the students who
scored 0 in the coding task (meanAL = .33, st.dev.AL = .23) and students who
scored 1 in the coding task (meanAL = .49, st.dev.AL = .31) (Table 4).
Table 4. Algorithmic task scores grouped by computational scores.
COM score AL mean AL st.dev.

0.0 0.33 0.23
0.5 0.40 0.26
1.0 0.49 0.31
4.3 School and Class Effects

School and class effects are analysed using a repeated measure ANOVA with 2
factors:
school (SCU): 1 – middle school, 2 – high school;
class (CL): 1 – first year, 2 – second year, 3 – third year.
Tasks main effect. The ANOVA shows there are main effects between the tasks
(F (1, 1785) = 53.12, p < .001), meaning that at least one of the tasks has sig-
nificantly lower scores than the others, independently of the school and class.
Figure 1 shows that there are drops in the mean scores obtained in spatial ori-
entation, temporal order and best solution recognition tasks.
School main effect. There is also a main effect of the school (F (1, 255) =
76.27, p < .001), meaning that there exists a significant difference between the
global score obtained from middle school students and the global score obtained
from high school students. A post hoc comparison confirms the significance of
this difference (pbonf < .001), and shows that middle school score (mean =
.65, st.dev. = .15) is lower than high school score (mean = .81, st.dev. = .14).
Fig. 1. Tasks main effect
Class main effect. The ANOVA also shows a main effect of the class (F (2, 255) =
3.78, p < .05) but this effect doesn’t have a direct interpretation, as it considers
first year students of both middle school and high school in the same group.
For this reason we prefer to repeat the ANOVA on middle school data and high
school data separated.
The results of the ANOVA on middle school data show there is no main effect
of the class, so the differences between the mean scores of the different classes
are not significant.
The results of the ANOVA on high school data, instead, confirms the main
effect of the class found originally (F (2, 119) = 3.02, p = .05) and the post hoc
comparisons show the only significant difference (pbonf = .05) is between first
year student scores (mean = .78, st.dev. = .15) and third year students scores
(mean = .85, st.dev. = .13).
Tasks and school interaction effect. The interaction effect of tasks and school is
significant (F (7, 1785) = 3.07, p < .005), meaning that there is at least one task
where the difference between middle school and high school scores is higher than
the differences in the other tasks. Figure 2 shows the task where the difference
between middle school and high school scores is higher is spatial orientation (O).
Task and class interaction effect. The interaction effect of tasks and class is not
significant, nor in the case we consider all of the data, nor in the case we consider
middle school and high school data separately. This means that the trend of the
classes among the different task is the same for all the classes, as we can see in
Fig. 3.
School and class interaction effect. The interaction effect of school and class is
again not significant, meaning that trend of the classes among middle school and
high school are similar, as shown in Fig. 4.
Fig. 2. Interaction effect of tasks and school. SCU: 1 = middle school; 2 = high school
Fig. 3. Interaction effect of tasks and class. CL: 1 = first year; 2 = second year; 3 =
third year
Fig. 4. Interaction effect of tasks and class. SCU: 1 = middle school; 2 = high school.
CL: 1 = first year; 2 = second year; 3 = third year.
4.4 Gender Effect

It was possible to mark down the gender only for high school students, but the
numbers show there are only 7 female students in the high school sample, so this
number is not significant to make any analysis.
Although we don’t know the individual gender for middle school students,
we know the gender distribution among the classes, see Table 1. From this dis-
tribution, we can see that some classes have a balanced number of male and
female students, while the class 1A has a higher number of female students and
classes 2A and 2B have a higher number of male students. Since there is no
significant difference between the classes in middle school, we could analyse our
data considering the class 1A as a class with female dominance and the classes
2A and 2B as classes with male dominance, and see if there is a significant
difference between classes with different gender dominance. The independent
sample t-test shows there is no significant difference in any of the tasks, in fact
the mean results of the first part of the test are very close (Female dominance:
mean = .63, st.dev. = .15; Male dominance: mean = .65, st.dev. = .17).
5 Conclusion
The intention of this work was to establish a preliminary instrument that can
help to introduce computational abilities in the secondary school.
The results show that the main abilities underlying computational thinking
are temporal order, best solution identification, spatial orientation and reasoning
by analogies. The school is a factor affecting students’ performances so these
abilities improve over time, especially after following specific courses, as in the
case of third year high school students. The analysis on the gender suggests
there is no difference between female and male performances, but this result is
not significant as it was not possible to collect individual informations.
This preliminary work has brought some positive results, but has also high-
lighted the necessity of some improvements. Hence we have in mind to redesign
the test to enhance its effectiveness and identify suitable action to improve the
introduction of computational thinking in non vocational school curricula.
We also plan to work in cooperation with secondary school teachers in order
to find an effective way to introduce computational thinking skills in secondary
school. Moreover we intend to deepen our analysis of the effect of school back-
ground on the learning of programming and other computing subjects for non
vocational university students.
References
1. (2017). www.bebras.org
2. Ambrosio, A.P., Almeida, L.S., Macedo, J., Franco, A.H.R.: Exploring core cog-
nitive skills of computational thinking. In: Psychology of Programming Interest
Group Annual Conference 2014 (PPIG 2014), pp. 25–34 (2014)
puter literacy: Criteria for good tasks. In: Informatics Education-supporting Com-
putational Thinking, pp. 19–30 (2008)
4. Dagienė, V., Sentance, S.: It’s computational thinking! bebras tasks in the cur-
5. Erdogan, Y., Aydin, E., Kabaca, T.: Exploring the psychological predictors of
programming achievement. J. Instr. Psychol. 35(3) (2008)
6. Knuth, D.E.: Algorithmic thinking and mathematical thinking. Am. Math.
Monthly 92(3), 170–181 (1985)
7. Korkmaz, Ö.: The impact of critical thinking and logico-mathematical intelligence
on algorithmic design skills. J. Educ. Comput. Res. 46(2), 173–193 (2012)
8. Mannila, L., Dagiene, V., Demo, B., Grgurina, N., Mirolo, C., Rolandsson, L., Set-
tle, A.: Computational thinking in k-9 education. In: Proceedings of the Working
Group Reports of the 2014 on Innovation and Technology in Computer Science
Education Conference, pp. 1–29. ACM (2014)
9. Milić, J.: Predictors of success in solving programming tasks. Teach. Math. 22,
25–31 (2009)
10. Papert, S.: An exploration in the space of mathematics educations. Int. J. Comput.
Math. Learn. 1(1), 95–123 (1996)
11. Pasini, M., Solitro, U., Brondino, M., Raccanello, D.: The challenge of learning
to program: motivation and achievement emotions in an extreme apprenticeship
experience. In: Proceedings of 27th Annual Workshop of the Psychology of Pro-
gramming Interest Group - PPIG 2016, pp. 151–156 (2016)
12. Pasini, M., Solitro, U., Brondino, M., Burro, R., Raccanello, D., Zorzi, M.: Psy-
chology of programming: the role of creativity, empathy and systemizing. In: Vit-
torini, P., Gennari, R., Di Mascio, T., Rodrı́guez, S., De la Prieta, F., Ramos, C.,
Silveira, R.A. (eds.) MIS4TEL 2017. AISC, vol. 617, pp. 82–89. Springer, Cham
(2017). https://doi.org/10.1007/978-3-319-60819-8 10
13. Plerou, A.: Dealing with Dyscalculia over time. Icicte 2008, 1–12 (2014)
14. Plerou, A.: Algorithmic thinking and mathematical learning difficulties classifica-
tion. Am. J. Appl. Psychol. 5(5), 22 (2016)
15. Raven, J.: Raven progressive matrices. In: McCallum, R.S. (ed.) Handbook of
Nonverbal Assessment, pp. 223–237. Springer, Boston (2003). https://doi.org/10.
1007/978-1-4615-0153-4 11
16. Solitro, U., Zorzi, M., Pasini, M., Brondino, M.: Early training in programming:
from high school to college. In: Gaggi, O., Manzoni, P., Palazzi, C., Bujari, A.,
Marquez-Barja, J.M. (eds.) GOODTECHS 2016. LNICSSITE, vol. 195, pp. 325–
332. Springer, Cham (2017). https://doi.org/10.1007/978-3-319-61949-1 34
17. Solitro, U., Zorzi, M., Pasini, M., Brondino, M.: A “light” application of
blended extreme apprenticeship in teaching programming to students of mathe-
matics. Methodologies and Intelligent Systems for Technology Enhanced Learn-
ing. AISC, vol. 478, pp. 73–80. Springer, Cham (2016). https://doi.org/10.1007/
978-3-319-40165-2 8
18. Syslo, M.M., Kwiatkowska, A.B.: Informatics for all high school students. In:
Diethelm, I., Mittermeir, R.T. (eds.) ISSEP 2013. LNCS, vol. 7780, pp. 43–56.
Springer, Heidelberg (2013). https://doi.org/10.1007/978-3-642-36617-8 4
19. Webb, M., Davis, N., Bell, T., Katz, Y.J., Reynolds, N., Chambers, D.P., Syslo,
M.M.: Computer science in k-12 school curricula of the 2lst century: why, what
and when? Educ. Inform. Technol. 22(2), 445–468 (2017)
Solving Everyday Challenges
in a Computational Way of Thinking
Bernhard Standl(B)
Institute of Software Technology and Interactive Systems, TU Wien,

Vienna, Austria
bernhard.standl@ifs.tuwien.ac.at
Abstract. Over a decade, computational thinking (CT) has been in

the focus of educators and researchers in computer science. During this
period of time, the term has been developed in different ways, reaching
from a fundamental idea for finding a definition what problem-solving
in computer science is about to a very particular view that CT is a
required skill to code software applications. This paper presents results
of the Fulbright project coThink - Computational Thinking carried out at
the Missouri State University in Springfield, MO, USA which was based
on the research question: How can CT be utilized with computer science
algorithms for challenging real-life situations? As a result of a literature
review, a CT five-step problem-solving process aimed at improving stu-
dents’ awareness to handle everyday life situations was identified. It was
further integrated in classroom lessons, where it was applied at four stu-
dent groups and evaluated mixing qualitative (analysis of worksheets)
and quantitative methods (questionnaire) at a sample size of n = 75.
Results showed that students frequently discovered a good approxima-
tion to solve real-life challenges following computer science algorithms
but we also came to the conclusion to revise our problem-solving process.
Keywords: Computational-thinking · Problem-solving · Classroom

research
1 Introduction
Seymour Papert initially introduced the term computational thinking (CT ) in

programming computers with LOGO for K-12 in 1980 in his book Mindstorms
[11], but with problem-solving in mathematics in his mind. On the other hand
Papert’s problem-solving processes were influenced by Poyla’s work from the
1950 s in How to solve it [12] who described how mathematical problems can
be systematically approached and solved by students. The current movement in
CT was initiated when Wing introduced the term again in 2006 [17]. Despite
some similarities between Wing’s and Papert’s definition, Wing emphasized on
problem-solving and Papert focussed on using CT to forge ideas [11]. Still, both
didn’t necessarily involve a computer in the process [6,9]. Considering this, we
https://doi.org/10.1007/978-3-319-71483-7_15
Solving Everyday Challenges in a Computational Way of Thinking 181
identified in recent literature, CT is for the most part either understood as

umbrella term for learning coding concepts as e.g. loops, variables, recursions or
as a term describing a problem-solving approach as used in computer science.
Just as Nickerson et al. argued in [10]: Definitions of CT vary, there is, however,
only little consensus in a definition for CT and so far a common understanding
can only be identified in Wing’s view [19]: CT is the thought processes involved in
formulating problems and their solutions so that the solutions are represented in a
form that can be effectively carried out by an information-processing agent. Barr
and Stephenson argued in [2] referring to Wing [17], that CT includes seeking
algorithmic approaches. This underlines that algorithmic thinking is part of CT
and a term for composing different approaches, techniques, and systematics used
in computer science for solving problems. Considering Lee, who argued in [8] that
algorithmic thinking is a part of CT and further Barr, where CT is defined as
the seeking algorithmic approaches to problem domains [2] in this paper we are
focusing on the algorithmic part of CT. Even if the term CT is popular these
days, so broad and deep it has been developed in history over decades. However,
Tedre and Denning further described in [14] that some CT strategies can be
applied also in different contexts, but they also claim, that such transfer has
never been substantiated. Taking Yadav’s thought, that CT can be integrated for
everyday life challenges into account [21], this paper introduces an approach for
combining computer science algorithms with real-life applications (as e.g. finding
the shortest path to the school) within the theoretical framework of CT aimed
at increasing the students’ algorithmic problem-solving awareness by addressing
these questions: What is a suitable step-by-step computational-thinking problem-
solving process for finding algorithms in real-life applications? How can such
process put in practice classroom teaching at high-school? The paper is structured
as follows: First we are discussing related literature and propose a framework for
a CT problem-solving processes for real-life applications, then we are presenting
the integration of the process in classroom practice followed by an evaluation in
practice and close with a conclusion and an outlook for further work.
2 Background
2.1 Algorithms for Everyday Life Applications
Algorithms in computer science have been developed for computers but they also
can address human questions of everyday life. Our motivation for finding real-
life applications is based on Wing’s point of view, that techniques of computer
science can support solving problems in other areas of life [17] where students
identify learning as more authentic and has a strong relation to their own life.
A further reason for connecting algorithms to human living was also, that this
approach allows students to focus on the process of problem-solving beyond
coding in a programming language. Inspired by popular ideas by Christian and
Griffiths in Algorithms to Live By in [4] and scientifically confirmed in Algorithms
Unplugged by Vöcking et al. [15] we selected five algorithms for an application
182 B. Standl
in classroom practice: binary search, bin packing, minimal spanning tree, topo-
logical sort, and shortest path. Supposing, that the algorithms itself are known
to the reader of this paper, we are presenting here the algorithms with only
a brief description of the idea for an application with everyday life problems:
Binary Search: When searching a CD or a book among many in a shelf, stu-
dents develop a search strategy to find the item quickly. This leads to search
algorithms and to the binary search algorithm in particular. Bin Packing: By
asking how moving boxes can be filled up effectively with things from closets and
shelves, students should find out strategies to master this systematically. The
algorithms Next Fit and First Fit will describe the solution in computer science.
Minimal Spanning Tree: The challenge is to connect seven islands with each
other in building bridges. The distance between each bridge is known. Students
have to find out the shortest possible connection to each island connecting the
mainland. Solutions will lead to Prim’s and Kruskal’s algorithm. Topological
Sort: Students write down a to-do list for a day and try to bring this list into
order. By doing this, dependencies and preconditions will lead to the aimed
algorithm for topological sorting that is used at finding one possible topological
order. Shortest Path: Students work on a street map and try to identify the
shortest path from home to school. The idea of selecting these algorithms was,
to show in examples that solutions from complex problems in computer science
can be easily transferred to challenge real-life situations as well.
2.2 Our View on Computational Thinking

Beyond all differences in definitions for CT, we identified across almost all con-
tributions, that CT is about an approach for problem-solving in a way, that is
used in computer science identified as algorithm. Wing already has stated in her
refined definition of CT in [18], that CT is an approach for solving problems
that draws on concepts fundamental to computing. Further, Aho described in
[1] that the term CT includes algorithm-design and problem-solving techniques
that can be used to solve common problems arising in computing. This is similar
to Wolfram Math World’s definition of the term algorithm itself as a specific set
of instructions for carrying out a procedure or solving a problem. The question
arises, if there is an additional value in introducing CT or if CT is just a vague
and fashionable term paraphrasing procedures already included in the term algo-
rithm? In fact, narrowing the term CT to designing algorithms would be off its
underlying idea. Selby et al. explained for instance in [13] CT should include
the ability to think in abstractions, in terms of decomposition, algorithmically,
in terms of evaluation, and the ability to think in generalizations. Moreover, Xu
et al. described in [20] CT as a universal, not limited to computers, fundamen-
tal, not only for experts, provides mental tools for thinking processes, has rich
contents and is connected to any concept of computer science, and poses deep
scientific problems in providing a framework to ask meaningful scientific ques-
tions. As Yadav et. al. reminded in [21,22] that Wing’s initial paper [17] points
out that CT involves three key elements Algorithms, Abstraction, and Automa-
tion, the term CT has been grown since then to a variety of interpretations. For
Lee et al. [8] CT involves defining, understanding, and abstraction. Barr et al.
suggested in [2] CT involves the design of solutions, implementation of designs,
testing, running analysing, reflecting, abstraction, creativity, and group problem
solving. Grover and Pea [7] stated that CT should include among others abstrac-
tion, information processing, structured problem-solving decomposition as mod-
ularization, iterative recursive thinking, and efficiency. Again, for Lee et al. [8]
CT involves defining, understanding, and solving problems, reasoning at multiple
levels of abstraction, understanding and applying automation, and analysing the
appropriateness of the abstractions made. In reference to Voogt et al. [16] who
suggest, that ...we should not try to give an ultimate definition of CT, but rather
try to find similarities and relationships in the discussions about CT, outcomes of
our literature review resulted in a problem-solving approach which goes beyond
the skills computer literacy and understanding technology and underlines the
need for developing CT skills [5]. Beyond the actual problem-solving process,
Barr et al. suggested in [2] also a collection of dispositions a problem-solver
should hold, which will be included in our process bellow. Those are confidence,
persistence, handle ambiguity, deal with open-ended problems and communica-
tion skills in team-work. From teaching practice experiences, we know that the
teacher’s awareness of such dispositions at the students can be crucial for the
success of learning and added them as additional part of the process.
2.3 Problem-Solving Approach for Real-Life Challenges

The goal of our CT problem-solving process is, to improve the students’ aware-
ness to be able to take advantage of CT problem-solving techniques to solve real-
life problems. Therefore, we suggest an arranged step-by-step process for solving
a problem. The process includes besides elements as abstraction, decomposition,
and design of algorithm also understanding the problem at the beginning of the
process and testing the solution at the end of it. 1: Understand the problem
as whole and restate the problem to unveil new perspectives to support the solu-
tion process. Also, state clearly what should be achieved with the solution. This
part derives from Papert’s initial ideas on CT [11], referencing to Poyla’s first
problem-solving step First, you have to understand the problem in How to solve
it [12]. 2: Abstract a problem in a way that helps to solve it. If we had to
keep all the details in our heads, we could never get anything done. As we have
described above, abstraction is mentioned as a keystone of CT and Grover et
al. identified it as a core element, which differs CT from other types of thinking
[7]. 3: Decompose the problem to break a hard problem up into smaller, easier
ones. Decomposition involves finding structure in the problem and determining
how the various components will fit together in the final solution. Doing decom-
position well makes it easier to modify the solution later by changing individual
components, and also enables the reuse of components in solutions to other prob-
lems. This was already suggested by Wing in [17] as modularizing a problem to
make it tractable. We see the parts decomposing and abstraction are key steps
for preparing the actual problem-solving process. Abstraction is a step, where
parts found will be removed or reduced in its representation and decomposing is
184 B. Standl
a process, where the remaining parts will be separated and clustered. 4: Design
an algorithm to develop the step-by-step instructions for solving the problem.
Start from what already is known and work outward from there. Make a plan
how to approach to solve the problem. Selby et al. [13] relate to Wing [17] and
further describe this process as a heuristic reasoning to devise a solution. 5: Test
the algorithm whether a solution meets the criteria as testing and debugging as
used in software development and in the context of CT suggested by Brennan et
al. in [3] to smoothen not working parts of the solution. This process has been
planned by us so that it can be carried out in a linear fashion step by step for
solving the problem. As mentioned above, we enhanced this five-step problem-
solving approach, with these dispositions, as it was suggested by Barr et al. in
[2]: Confidence in dealing with complexity, Persistence in working with diffi-
cult problems, Tolerance for ambiguity, The ability to deal with open ended
problems, The ability to communicate and work with others to achieve a
common goal or solution. Taking this approach for a CT problem solving app-
roach into account, in the next section we will describe, how we integrated it
into classroom practice.
3 Method
In order to proof the concept, we carried out our interventions at two high-
schools with two classes each. The participants (n = 75) were students between
the ages of 15 and 17 from four different high-school classes at two differ-
ent schools (Greenwood Laboratory School and Willard High School located
in Springfield, MO, USA). As the access to high school classes was limited
to 2 × 2 h, we designed our lessons with a strong emphasis on your five-step
problem-solving approach. Our goal was to find out, if everyday life problems
can be approached systematically using techniques from computational think-
ing. Therefore, we planned four lessons on two days with the overall intention
to let students find a solution to a problem heuristically followed by a reflec-
tion compared to the actual algorithm of computer-science. Students not only
practiced their problem-solving competencies but also learned how computer sci-
ence concepts can be useful and meaningful for their own life. We accompanied
the classroom activities with mixed methods using a pre-/post-questionnaire
and an analysis of the students’ work sheets. Even though, both schools offer
computer science classes with instruction on using a computer and standard
software, students of both schools received no prior computer science education
with a focus on computational algorithmic thinking on a regular basis. Hence,
we presumed that all participating students had similar pre-experiences with
algorithmic problem-solving processes. Due to this fact, we identified all stu-
dents across both schools as one entity for analysing the data. It should also be
mentioned that we did not include any control groups for organizational reasons
and because of the research design, which was primarily aimed at evaluating
our problem-solving concept. We carried out two separate lessons (2 h each) on
two different days with each of the four student groups at two schools, the first
lesson introduced CT and our problem-solving approach and in the second les-
son students worked on problems individually and in groups collaboratively. In
order to provide students a guideline through the problem-solving process, our
worksheets were distributed with a clear description of the problem and a path
through the process which was the same as stated above from understanding the
problem for generalizing the solution.
3.1 Instruments
We designed two research instruments for evaluating our problem-solving

process: in order to track the students’ problem-solving strategies, we designed
worksheets. The worksheets had five sections, divided into the five-step problem
solving process. In Fig. 1, the first two pages of a worksheet to bin-packing are
depicted. The front page describes the problem to the student and recalls the
five-step problem-solving process and the supportive attitudes in the column on
left hand side. On the right-hand side, an exemplary part of the work-sheets
shows, that students also had to sketch for each problem-solving step their ideas
as mentioned above: describe the problem, abstract the problem, decompose
the problem, Design the algorithm, test the solution. In order to identify differ-
ences in students attitudes towards problem-solving we distributed pre-/post-
questionnaire. The questionnaire included 20 items, where each of Barr’s dis-
positions [2] has been assigned to four questions using a Likert-Scale from 1
‘strongly disagree’ to 6 ‘strongly agree’.
Fig. 1. Worksheet for the task “Pack your moving boxes”

186 B. Standl
3.2 Analysis
The analysis of the worksheets was a process of interpretation, where we eval-
uated our linear approach of solving the problem in going through each of the
five problem-solving steps. We checked in which parts of the worksheets students
were accepted as useful during the problem-solving process by reading through
the worksheets and identified, which parts were not edited by the students where
it was difficult for students to use it for the problem-solving process. The analy-
sis of the questionnaire was carried out with descriptive statistics and statistical
significance tests.
3.3 Classroom Lessons

The lesson plan in Fig. 2 gives an overview on the classroom intervention of day
one and is based on findings of our literature research and includes all five steps
from understanding the problem to generalizing the solution.
Fig. 2. Lesson plan day 1 and 2
We started with an introductory lecture about problem-solving, CT and algo-

rithms. Next the teacher approached a solution together with the whole student
group for the shortest path problem by going through the five-step problem solv-
ing approach introduced above. Next, each student solved the example with the
minimal spanning tree and shared the solution at the end of the hour together
with others in group-work. It is important to note, that the students’ attitudes
required for a beneficial problem-solving process were mentioned explicitly by
the teacher, who facilitated also the group-work for promoting the students’
awareness for it. On the second day, the teacher first summarized insights from
the first lesson and gave a lecture on some approaches, how computer scientists
try to solve problems. This included basic sorting algorithms as well as general
strategies as divide-and-conquer. Students subsequently worked in groups on
more difficult problems as binary search, bin packing and the topological search
as described above. Finally, group-work was followed with group presentations
which included a discussion on connecting solutions with the real algorithms
from computer science.
Example of application: In order to underline how our CT problem-solving
process works for real-life challenges in detail, we give here an example based on
the worksheet as depicted above. To each problem-solving section, we describe a
possible student’s answer which is based on Vocking et al. [15]. Pack your mov-
ing boxes: How can I pack moving boxes effectively? Possible solution of a
student: By going through the problem-solving process, the students systemat-
ically approach the solution to the problem (very short version): Describe the
problem: The challenge is to place my items into moving boxes in the most
efficient way. Abstract the problem: To make it easier, I reduce the items
to a geometrical representation that each part is described as block. So, a ball
becomes for example a block with a edge length of its diameter. Decompose
the problem: I have moving boxes with a certain size and items with different
sizes and forms. Design the algorithm: First I tried different approaches in
practice and extracted a general solution: 1. Sort the items by size, 2. Take the
next available biggest item. 3. If the items fit into the box: Place the item into the
box. Else: Open the next box. 4. Go to step 2. Test the solution: I organized a
small shoe-box and some items and I evaluated my solution. It turned out, that
my solution is not perfect yet as I need to many moving boxes. This example
demonstrated, how the problem-solving process can be integrated for finding a
solution to a problem from a real-life context by using a CT problem-solving
approach leading to an algorithmic solution.
4 Results
4.1 Worksheet: Problem-Solving Process
Results showed, that students were successful in developing an approach to solu-

tions for the problems. But they also experienced sometimes difficulties in dealing
with the five-step process of solving a problem and were not always sure how
each of the steps relates to the problem-solving process. For example, students
were more engaged in describing the problem and designing the solution but
they had difficulties to fill the worksheets for abstracting the problem, decom-
posing it and evaluating the solution separately. We found out, that the process
of describing the problem already covered parts of the abstraction process. As in
Fig. 3 it is shown, students have used different ways to solve the problem, some
graphical, some using text. For instance, when students described the problem
using their own words, some kind of abstraction has already taken place by most
of the students. Next, the part of decomposing the problem in smaller parts apart
188 B. Standl
from problem-solving was experienced as difficult and was rarely edited. Most of
the time, decomposition was integrated into the design process or the evaluation
process. These outcomes suggested us to reorder and reduce our problem-solving
process at the worksheets in future to three steps: 1. Describe, decompose
and abstract the problem, 2. Design the algorithm, and 3. Test the
solution.
Fig. 3. Snippets of filled student worksheets
4.2 Questionnaire: Students’ Problem-Solving Attitudes
Even though, the structure of the worksheets was challenging for students, our
intervention still had some impact on students’ attitudes towards computational
problem solving as the pre-/post-questionnaires suggest. The results of the ques-
tionnaires showed some differences at certain items, where some of them even
differed statistically significant. The items of the diagram below can be clustered
in four sections, where each cluster represents a field of dispositions a problem-
solver should hold as defined above. These are 1–4: Confidence in dealing with
complexity, 5–8 Persistence in working with difficult problems, 9–12: Tolerance
for ambiguity, 13–16: The ability to deal with open ended problems, 17–20:
The ability to communicate and work with others to achieve a common goal or
solution. The questionnaire was identified as reliable as all two of the measures
evidenced a good reliability as Cronbach’s alpha ranged from 0.82 for the pre-
test and 0.85 for post-test. As the results in Table 1 show, confidence (items 1–4)
increased and item 1: “I feel anxious when I have to solve this kind of difficult
problems.” had also a significant difference in the pre-test scores (M = 3.55, SD
= 1.51) and post-test scores (M = 4.20, SD = 1.26); t(140) = −2.77, p = 0.006.
Furthermore we identified a significant difference in the pre-test scores (M =
3.55, SD = 1.42) and post-test scores (M = 4.20, SD = 1.12) at item 4: “I feel
Table 1. Aspects from students’ attitudes towards problem solving. Scales from 1
‘strongly disagree’ to 6 ‘strongly agree’ (n = 75)
Item ∅ pre ∅ post

1 I feel anxious when I have to solve this kind of difficult problems 2.78 2.00
2 I doubt that I could solve this complex problem 2.26 1.73
3 I like to solve challenging problems like this one 3.74 3.55
4 I feel confident in solving this kind of difficult problems 3.70 3.91
5 If I could not solve this problem quickly, I would give up 2.30 2.18
6 I feel impatient when I’m working on this kind of challenging problems 2.87 3.09
7 When a problem like this challenges me, I am keen to solve it 4.17 4.09
8 I would work on this problem until I have found a solution 4.39 3.73
9 I rather prefer to work on problems where the solution is already obvious 3.04 3.09
10 There is possibly only one way to solve this problem 2.70 1.91
11 There are many roads that lead to the solution 4.13 4.64
12 This problem may has more than one solution 4.61 4.73
13 There exists a solution to this problem 5.04 5.55
14 I would be confused, if I wouldn’t find a clear solution with this problem 3.78 3.82
15 There can be different approaches to solve this problem 4.78 5.00
16 The solution to this problem can look different from what I have expected 4.43 4.82
17 I don‘t think that someone could help me to solve this problem 2.13 2.27
18 If too many people are involved in solving this problem, it will not be done well 3.43 3.95
19 Teamwork is the key to solve this problem effectively 3.78 3.93
20 A solution to this problem could be better found together 4.26 3.27
confident in solving this kind of difficult problems.”, conditions; t(140) = −3.00,

p = 0.003. Items in the second cluster of persistence in the problem-solving
process (items 5–8) decreased. The third cluster tolerance for ambiguity (items
9–12) increased and there was a significant difference in the pre-test scores (M
= 4.06, SD = 1.14) and post-test scores (M = 4.74, SD = 0.89) at item 10:
“There is possibly only one way to solve this problem.”, conditions; t(140) =
−3.59, p = 0.001. The fourth cluster on the ability to deal with open ended
problems (items 13–16) increased and there was a significant difference in the
pre-test scores (M = 4.94, SD = 1.51) and post-test scores (M = 5.34, SD = 1.26)
at the variable “There exists a solution to this problem.”, conditions; t(140) =
−2.32, p = 0.022. The fifth cluster on teamwork (items 17–20) is inconsistent in
terms of that students think that the solution to a problem can be found better
together (item 20) but at the same time state that too many people involved in
problem solving will have a negative impact (item 17). In a nutshell, results of
the questionnaires suggest that students reduced their anxiety in solving diffi-
cult problems and increased their confidence. We further conclude, that students
developed their understanding for the possibility of multiple solutions.
5 Discussion
This paper presented results of the project coThink - Computational Thinking
which was aimed at identifying an approach for a step-by-step CT problem-
solving process for real-life applications. The underlying idea was to break
190 B. Standl
problem-solving techniques known from computer science down to support stu-

dents’ awareness for handling real-life challenges. Based on literature research
a five-step CT problem-solving approach emerged which was enhanced with
dispositions a problem-solver should hold as suggested by Barr et al. in [2].
We designed lessons and material which integrated our ideas and concepts and
applied it in practice at four student groups at two different schools. First expe-
riences showed, that students were keen to develop solutions for problems in a
real-life context and their approach mostly approximates an algorithmic descrip-
tion or even sometimes an existing algorithm of computer science. Results of the
questionnaires evaluating students’ attitudes for problem-solving showed, that
our intervention had some positive impact, even if it is likely that students didn’t
adopt their dispositions but rather had an increased awareness which disposi-
tions are required for CT problem-solving. Further work will also include the
refinement of new worksheets, the specification of further algorithmic real-life
examples and its application in classrooms. It turned out, that students had
problems filling all sections of the worksheets and finding a description for each
of the five problem-solving steps was perceived as confusing. In fact, only the
sections Understand the problem and Design the solution were accepted and
helpful in the process. Some of the students described their solution, which
was supposed to part of section Abstract the problem, as part of Describe the
problem. As a consequence, we further redesigned the worksheets and combined
fields which will lead to three sections of the worksheets, as described above (1.
Describe, abstract and decompose the problem - 2. Design the algorithm - 3.
Test the solution) In essence, we reduced the five-step problem-solving approach
to three steps in order to simplify the process for students to: understand - solve
- analyse. In the future, we will continue refining the teaching material and the
research setting for a more detailed analysis focusing more on the actual stu-
dents’ problem-solving process by adding for instance think-aloud interviews as
research instrument to gain more detailed insights how our approach has impact
on CT problem-solving process.
Acknowledgments. This research was supported by the Fulbright Visiting Scholar

Program Austria sponsored by the U.S. Department of State. I would like to thank the
Fulbright Office in Vienna, the Council for International Exchange of Scholars (CIES)
and the Institute of International Education (IIE). I thank my colleagues at Missouri
State University in Springfield, MO, USA who provided assistance in facilitating the
research project. I would also thank the school authorities and teachers at Greenwood
Laboratory School and Willard High School for support and assistance in organization
of classroom lessons and in submitting the IRB proposal.
References
1. Aho, A.V.: Computation and computational thinking. Comput. J. 55(7), 832–835
(2012)
2. Barr, V., Stephenson, C.: Bringing computational thinking to K-12. ACM Inroads
2(1), 48 (2011)
3. Brennan, K., Resnick, M.: New frameworks for studying and assessing the devel-
opment of computational thinking, pp. 1–25 (2012)
4. Christian, B., Griffiths, T.: Algorithms to Live by: The Computer Science of Human
Decisions. Henry Holt and Co., New York (2016)
5. Dagiene, V., Futschek, G.: Bebras, a contest to motivate students to study com-
puter science and develop computational thinking. In: Proceedings of WCCE, pp.
139–141 (2013)
6. Dagienė, V., Sentance, S.: It’s computational thinking! Bebras tasks in the cur-
7. Grover, S., Pea, R.: Computational thinking in K12 a review of the state of the
field. Educ. Researcher 42(1), 38–43 (2013)
8. Lee, I., Martin, F., Denner, J., Coulter, B., Allan, W., Erickson, J., Malyn-Smith,
J., Werner, L.: Computational thinking for youth in practice. ACM Inroads 2(1),
32 (2011)
9. Mannila, L., Dagiene, V., Mirolo, C., Settle, A.: Computational thinking in K-9
education, pp. 1–29 (2014)
10. Nickerson, H., Brand, C., Repenning, A.: Grounding computational thinking skill
acquisition through contextualized instruction. In: Proceedings of ICER 2015, pp.
207–216 (2015)
11. Papert, S.: Mindstorms: Children, Computers, and Powerful Ideas. Basic Books,
Cambridge (1980)
12. Polya, G.: How to Solve It, 2nd edn. Penguin Books Ltd., London (1957)
13. Selby, C.C., Woollard, J.: Computational thinking: the developing definition, pp.
5–8 (2013)
14. Tedre, M., Denning, P.J.: The long quest for computational thinking. In: Koli
Calling Conference on Computing Education Research, pp. 120–129 (2016)
15. Vöcking, B., Alt, H., Dietzfelbinger, M., Reischuk, R., Scheideler, C., Vollmer, H.,
Wagner, D. (eds.): Algorithms Unplugged. Springer, Heidelberg (2010). https://
doi.org/10.1007/978-3-642-15328-0
16. Voogt, J., Fisser, P., Good, J., et al.: Computational thinking in compulsory edu-
cation: Towards an agenda for research and practice. Educ. Inform. Technol. 20,
715–728 (2015). https://doi.org/10.1007/s10639-015-9412-6
17. Wing, J.M.: Computational thinking. Comm. ACM 49(3), 33 (2006)
18. Wing, J.M.: Computational thinking and thinking about computing. Philos. Trans.
Ser. A Math. Phys. Eng. Sci. 366(1881), 3717–3725 (2008)
19. Wing, J.M.: Computational Thinking: What and Why? The Link - The Magazine
of the Carnegie Mellon University School of Computer Science (2011)
20. Xu, Z.W., Tu, D.D.: Three new concepts of future computer science. J. Comput.
Sci. Technol. 26(4), 616–624 (2011)
21. Yadav, A., Hong, H., Stephenson, C.: Computational thinking for all: pedagog-
ical approaches to embedding 21st century problem solving in K-12 classrooms.
TechTrends, 1–4 (2016)
22. Yadav, A., Zhou, N., Mayfield, C., Hambrusch, S., Korb, J.T.: Introducing compu-
tational thinking in education courses. In: Proceedings of the 42nd ACM Technical
Symposium on Computer Science Education, pp. 465–470. ACM (2011)
System Papers
Teaching Basic Elements of OOP in School
Informatics During Constructing Virtual
Micro-worlds
Evgeny A. Eremin(&)
Perm State Humanitarian Pedagogical University, Perm, Russian Federation

eremin@pspu.ru
Abstract. Object-oriented programming (OOP) is one of the most prevalent

techniques of programming now. OO principles are also widely used in com-
prehensive software including user graphical interface. Although the benefit of
early teaching OO concepts is advisable, it meets variety of difficulties and as a
result no general consensus on this subject exists. The present paper by virtue of
OOP instruction analysis proposes to include OO foundations as a topic into the
school informatics course. The specific of the suggested method lies in a sep-
aration of OO principles from learning any concrete programming language like
Java or C#: the original software tool was developed for these aims by the
author. The tool acquaints pupils with OO concepts during construction from
objects their own virtual world.
Keywords: OOP Objects School Informatics Courseware

Micro-worlds
1 Introduction
At present an object-oriented approach became the prevalent paradigm in the profes-

sional world of software development. Programming community recognized the value
of OOP for designing complex systems and now practically all modern programming
languages include support of this technology. Objects are widely used not only for
programming, but also for internal organization and interfaces in many software sys-
tems. For instance [16] OO approach helps to understand the notion of process in a
multitasking system, the message exchange mechanism and even user manipulations
with the standard elements of graphic interface. The ideas similar to OO inheritance can
be found in artificial intelligence, where hierarchical structures represent human
knowledge. The architecture of modern computers is formed from modules with very
multifarious internal organization, connected by a common interface. Besides in edu-
cation we can see that computer-based courses and courseware are often constructed
from modular objects [26].
Such wide application of OO ideas in industry, computer science and engineering
education requires its wide learning, but this demand comes across many difficulties.
First of all OO approach is based on a high-level abstraction, so understanding of this
subject is a serious cognitive problem [9, 10, 23, 24, 29]. Furthermore it is usually

https://doi.org/10.1007/978-3-319-71483-7_16
196 E.A. Eremin
connected with a simultaneous learning of programming language constructions. Some

other plural subjective factors also complicate OOP teaching [13], for example, an
absence of suitable well-developed teaching materials and software tools. After an
extensive survey of literature on teaching OOP in [20] the authors also made a con-
clusion about a lack of provable pedagogical approaches, appropriate books and
suitable environments for novices. Although many ideas exist how OOP should be
taught to beginners, there is no general consensus on this subject.
This paper proposes to begin learning the fundamental ideas of OO approach in a
school informatics course. A new instructional method for these aims is described,
which helps to form the main OO concepts in pupil’s mind during building some world
on a computer display. Software support for this learning activity was written by the
author. For those who wish to create more advanced versions of computer support the
paper contains a description of the software system organization; using it, a pro-
grammer can develop more perfect educational software.
2 Background
“It is clear that novice programmers face a very difficult task. Learning to program
involves acquiring complex new knowledge and related strategies and practical skills”
[22]. Learning OOP is much more complicated goal [23]. So in this section we discuss
the difficulties of teaching OOP and the ideas how to overcome them. As the number of
publications about OOP is enormous, only the most related ones are mentioned in the
review.
2.1 Teaching Objects First

An experience of teaching OOP immediately showed that it has specific whether
student has got preliminary knowledge in “classic” (imperative) programming or not.
The phenomenon when learners meet difficulties, studying OOP after other non-OOP
languages, got a special name “paradigm shift”. To avoid this obstacle, a teaching
technology called “objects-first” was proposed. As follow from the name, it suggests
introducing OOP principles from the very beginning. So students start from objects and
inheritance, and after experimenting with these ideas only then “goes on to introduce
more traditional control structures, but always in the context of an overarching focus on
object-oriented design” [1].
Situation with this educational technology is not fully clear yet. An all-around
discussion about “objects-first” and “objects-late” technologies (“object vs. impera-
tive”) had not leaded to unity and consensus [14]. Psychological studies of a problem
are rare and non-systematic, as review [23] argues. Furthermore the experimental
research also brings dissonant results. For example publication [7] states the following.
“The main result of the study is that there are no differences between OOP-first and
OOP-later with regard to learning gain.” Conversely, research in [28] leads to the
positive conclusion: “it is possible to say that the learners instructed with object-first
method achieved higher learning outcomes.”
Teaching Basic Elements of OOP in School Informatics 197
It is worth mentioning that experimental studies clearly show the difference in

understanding of the program texts [21]. “Novice programmers comprehending an OO
style program form a strong domain model, while novices comprehending an imper-
ative style program form a strong program model”. More detailed review of this subject
can be found in [22]. From our point of view this thesis confirms that OOP under-
standing has higher level and so it is one more argument to learn it.
2.2 Avoiding Misconceptions of Terms

Another problem of OOP teaching is possible misconceptions about the fundamental
terms. A large set of such confusions were revealed after the analysis of the distance
learning experience in the paper [9]. As it turned out, students often mix even two basic
concepts – “object” and its “class”. Later publications [20, 24, 29] confirmed this
misconception.
What pedagogical conclusion follows from this surely established fact? The fun-
damental OOP concepts that do not depend upon any programming language must be
taught very carefully. Probably it is rational to learn them apart from a real pro-
gramming language.
2.3 Language Choice; Micro-worlds

Teachers very often debate the topic what language is better to instruct programming.
Researchers also persecute many studies making the detailed comparison of different
programming languages [15, 17]. A discussion about OOP language for the beginners
even more actual because the novices “seem to fixate on language details, losing the
big picture, no matter how much we stress all of these concepts” [24].
It is evident that a language for the novices must be laconic, with simple rules and
without unnecessary facilities. Such mini-languages [5] are often specially created for
learning of the programming foundations. “The time required to master a
mini-language itself is small, so the students can spend most of their efforts on more
important issues” [5].
Except simplicity, educational mini-languages have another essential feature: they
are always built on the base of some virtual environs, so a student’s program controls
certain object (actor). Such teaching environments are usually called micro-worlds [1,
5, 17, 27]. Hence using a micro-world in instruction has two essential advantages: the
motivation of pupils (introducing elements of game) and the reduction in complexity of
a programming language and a development tool.
There are several well-known micro-worlds to study OO approach, for instance
Alice [6], Greenfoot [11] and ObjectKarel [30], built on a base of Karel++ [3]. All of
them are successfully using as an introduction into one of the most popular OOP
programming language Java.
2.4 Visualization
Changes in a micro-world on a computer screen demonstrate the process of a program
execution, so it clearly visualizes what program a student has written and how accurate
198 E.A. Eremin
it is done. Modern professional programming environments include well-developed

tools for visualization. As it is argued in [12], for introductory courses the environment
is possibly more important than the language, because if it is too complex, pupils may
loose a sense of learning material, struggling with powerful, but unclear interface.
We must carefully examine what kind of visualization our students need. “The most
advanced professional object-oriented development environments, such as Visual C++
or Delphi, use graphical support only to build the user interface of an application, but
neglect the internal structure of the program itself” [12]. Besides educational software
for learning OOP need appropriate visualization of the basic OO concepts.
For further consideration it is important not to miss one more positive feature of the
present-day visual software systems: they allow us to get rid of writing numerous
formal declarations. When, using a mouse, you create a new component in program-
ming environment like Delphi, the system automatically generates necessary descrip-
tion and places it into a program text. The present paper expands this pupil friendly
idea: it suggests hiding all formal descriptions inside a teaching system and showing
them only in the demonstrative visual forms like tree of classes or tables of properties.
2.5 Application of General Didactic Principles

Lastly let us look at a problem of OOP teaching from the general pedagogical positions.
From this point of view we have a complex learning material that is difficult to teach.
What theory of didactics offers us for these aims? We can use so called “spiral
curriculum”. In his well-known book [4] Bruner wrote about studying the scientific
foundations at school the following way. “Experience… points to the fact that our
school may be wasting precious years by postponing the teaching of many important
subjects on the round that they are too difficult.” And later: “the early teaching of
science … should be designed to teach these subjects with scrupulous intellectual
honesty, but with an emphasis upon the intuitive grasp of ideas and upon the use of
these basic ideas. A curriculum as it develops should revisit these basic ideas repeat-
edly, building upon them until the student has grasped the full formal apparatus that
goes with them.” As we see from discussed above, OOP, being an approach that is built
on several non-language key concepts, is a good candidate to such scheme of learning.
Bergin in his project “Fourteen Pedagogical Patterns” [2] declared similar ideas
about building an educational course. “Organize the course so that the most important
topics are taught first. Teach the most important material, the “big ideas,” first (and
often). When this seems impossible, teach the most important material as early as
possible” [2]. The argument for such conclusion is that if you delay the learning of
important topics students may get many misconceptions (see segment 2.2 about
problems with fundamental OOP terms).
Applying these general principles to introducing OOP into a school course of
informatics, we can conclude that the starting point for it is not the constructions of
some popular programming language, but understanding what an object, its properties
and behavior is. Supported with a suit of exercises aimed on manipulation with objects
of some micro-world, such lessons will create the solid platform for future
programming.
3 Teaching Strategy
The analysis in the previous section showed that for many reasons it is practical to
begin an introduction into OO approach from teaching several fundamental categories
like object, class, inheritance etc. This part of material has no direct connection with
any OOP language. If we support it by some OO micro-world, we can get a complete
topic to learn within a framework of school informatics.
To determine the content of a learning material for this topic we may follow the
publication [31], based on authors’ programming and pedagogical experience, where
the six-step approach to teach the OO methodology is presented. First five steps are the
following (the sixth one is optional and does not take out below).
1. Discuss fundamental principles of object-orientation with respect to conventional
thinking.
2. Introduce an object concept by observing the real world.
3. Acquire the class concept by abstraction of many common objects.
4. Introduce instantiation after the class concept is learned.
5. Illustrate subclasses by adding more details to an existing class and superclasses by
finding common things among several classes.
First two steps are general and so should be discussed with the help of illustrations
from everyday life. For steps 3–5 the authors suggested to show students (as an
example) how correspondent declarations look in C++ language. But an example set is
not mandatory, so we can consider that all five steps do not depend upon specific
computer language. Temporarily compromising learning goals and working on them in
isolation is the main way to reduce cognitive load [25].
“After these steps, instructors can start the language teaching, and should be
confident that the students can learn the language with correct thinking and with
object-oriented methodology in mind” [31].
Similar strategy we find in [19] before an introductory course of Java language. It is
reported shorter, but the content is obviously similar. The first lectures explain the object
ideas using real world examples to form “mental models” of classes and objects. The-
oretical part are followed by several specially constructed for the course computer labs.
In recently printed textbook [18] such way to start OOP is described the following
way. “For some time you will not feel that you are learning programming. You will
have a feeling that we are only playing with the computer. However, in this initial
phase you have to learn a lot of terms, connections and regularities so that later on you
could understand further topics much better and quicker.”
So we see that an easy method to introduce OOP basics to novices is possible and its
content is clear. Now we shall describe its software support, proposed in the present paper.
4 Software Support
To support the theoretical explanations what is an object and how to use it, we should
have some learning software to organize labs. In this section a variant of such soft-
ware, proposed by the author, will be briefly described. It was named System Builder
200 E.A. Eremin
(SB) – a tool to construct a virtual micro-world from objects. First version of this
program was presented earlier [8], later on it was extended and improved after testing
with the students of pedagogical university (recent pupils and future school teachers).
4.1 Tool’s Peculiarity

As it was mention above, the most known micro-worlds are aimed at further learning
Java. Unlike them, SB is separated from concrete programming language and is
focused on the fundamental OO concept themselves. It is rather a tool for object design,
which teaches pupils to see objects with their relations in real life and build an OO
model of a situation. During manipulations with created objects (mainly with visual
images) pupils can deeper understand what is property, method, inheritance, poly-
morphism etc. On our opinion, such approach is right for school introductory topic.
Extending ideas of modern visual programming environments, SB has no any
descriptions of classes and objects at all. For instance to create a new type of objects
(class) you just fill an “electronic blank”: input its name into a text field, select a parent
class from a list and enter optional comment for yourself (Fig. 1). So we just give
answers to short clear questions instead of writing long statements, constantly
remembering rules and agreements of language. The system stores all input data into
memory and represents it for learners in a form of tables. Table representation, as it
turned out, is not only a demonstrative form of output, but also a perfect instrumen-
tation for compilation of a project.
According to its pedagogical aims SB has the easiest language to describe the
behavior of objects. It contains simple assessment (e.g. x.image = x.image + 1) and
calling of object’s method (x._changeimage). A conventional form of if-operator (with
possible nesting) is defined. In addition several optional constructions exist (e.g. a cycle
Fig. 1. Creating a new type of objects (class) from one of the standard types
for sequential substitution of objects with specified type in order, taken from corre-
spondent table). All these operators are used to write methods and method’s body
(without any formal title) is the longest text student must type. Minimal text with
primary intuitive syntax, fully free of descriptive statements from a concrete profes-
sional language (like import, void, static, public etc.), is an essential advantage of the
proposed teaching tool.
For maximum reducing of input, SB has predefined library of types, properties and
methods. It is read from a system text file when SB starts. For instance, there is a
primary hierarchy of objects in this library.
All object types finally are derived from the top parent type called AbsObject
(abstract object). Its descendants are MainObject with predefined object World and
Vis-Object, which describes a static (motionless) object with image. The last one’s
child named MovObject in addition can move (see hierarchy in Fig. 1). So a pupil can
instantly start to create visual static or moving objects.
4.2 How to Build a Virtual World

After carefully planning of their world, students run SB and do the following steps.
1. Construct the world: define object types, methods and properties of future objects.
If the standard template is not enough, add new types, properties and methods to it.
Try to prepare the maximum, although SB allows correction at any moment. Note
that as practice showed just this step often puzzles students unlike two further ones.
2. Build the word: using what was prepared on the previous step, create objects and
set properties to each of them. On this step a pupil already sees where objects are
situated and what images (static yet) they have.
3. Watch what was built. At last watch the evolution of world in time. If there are no
mistakes, enjoy how your ideas become true. If it is not so – try to find mistakes. SB
has special regimes to facilitate this process. Except automatic regime, there are two
debugging ones: by time steps and by operations.
4.3 Examples of Projects

Primarily let us briefly discuss as an example how to build in SB a simple project
named Timer (that shows minutes and seconds). Its view you can find in the top right
corner of the collage compounded in Fig. 2.
The sequence of operations may be the following.
1. First of all draw (or take from a clipart) 10 pictures with images of numerals, name
these files ‘pic1’ – ‘pic10’ and put them into a project folder.
2. Create a new type (called for example digit) from VisObject standard class and add
the only property max_value to it (see step 6 also). Let us accent, that timer digits
have no special property to keep current numbers – their images do this.
3. Add property carry to the standard object World. When any digit reaches max_-
value, it must be forced to 0; in this case carry = 1, otherwise carry = 0. (We can
see the simplest variant of data exchange between objects in this project.)
202 E.A. Eremin
4. Find in the standard type MainObject method _main and complete it with the text
from listing 1.
5. Add method _inc to digit type and input its program from the same listing. (Note
that this is the longest text to input.) Now our world is constructed.
6. Then create 4 objects of digit type (4 timer digits): sec_lo, sec_hi, min_lo and
min_hi. Postfixes _lo and _hi let us distinguish the low and the high digits in
minutes and seconds values. For both high digits max_value property must be set to
5 and for the low ones – to 9. Now all preparations are over, the world is built,
7. Run SB project and watch how timer becomes alive. Correct errors (if any).
Fig. 2. Examples of object projects, developed in SB

Short notification with cycle for in method _main is not mandatory; instead of it
you can write a longer variant with four similar fragments – for every digit singly.
Omitted name of an object in _inc listing is equivalent to this construction in Java.
It is necessary to emphasize that we have discussed one of the simplest static
projects. Much more complex ones, dynamic and interactive, are also available in SB
kit set as demo projects. In Fig. 2 several example are combined together in the collage:
• Plane – driving a plane through thundery clouds, using arrow keys (interactive);
• Traffic – crossroad controlled with a traffic light;
• Timer – timer with minutes and seconds;
• Lift – actor goes up and down by elevators, executing its program;
• Logic – digital logical scheme (interactive);
• Polymor – illustration that different types of objects may travel different ways;
• Neko – famous animation, made by Kenji Gotoh.
4.4 How it Works

SB keeps values of all properties for all objects as a matrix. For instance program line
World.timer = World.timer + 1 has internal representation 1.12 = 1.12 + #1, which
adds direct value 1 to property 12 of object 1.
According to matrix values, visual machine draws the window with our virtual
world (Fig. 3). To change a picture we must call any method, which assigns new values
to the properties and this will be immediately indicated on a screen. To make our world
“alive”, SB has timer (its frequency can be adjusted by users). At every tick of this
timer method _main is called. It does some actions according to our program, changing
the values of properties or calling some methods to do it. Finally the visual machine
repaints a screen and the process repeats again.
If we press some key or click a mouse button, SB starts one of the special system
methods with predefined name, also changing our virtual world through the values of
modified properties. This way our virtual world becomes interactive.
204 E.A. Eremin
virtual world visual

machine
user methods timer user properties
_main
_@up _@click random
_@enter
methods properties
Fig. 3. Functional scheme of the software
5 Conclusion
The analysis made above, confirms that there are convincing reasons to teach OO
basics in a school informatics course. Present paper proposes to do it out of direct
connection with concrete professional programming language like Java or C++. The
matter of the suggestion is to learn this topic during construction of a virtual world from
visual objects of different types and describing its behavior by means of the easiest
language. The special educational software was developed for these aims; it allows
building systems by writing minimal number of program lines.
Easy realization of OO project in SB is achieved by several means. Primarily it is
the wide usage of graphical interface. Every object, property or method is creating with
the help of “electronic blanks” – the dialog windows with standard graphic controls
like text fields, dropdown lists and so on. Instead of writing long statements constantly
digressing on syntax rules a pupil just answer clear questions and the software con-
structs components of OO world automatically. Besides all data about objects is stored
and displayed in a form of tables.
Although this paper describes the given author’s free software, its main aim is to
demonstrate possible way of teaching basic OOP ideas at school. Practice distinctly
indicates that students, working with SB, have the ultimate difficulties just when they
build an object model but not in the following implementation. So from this point of
view the reported method seems very advantageous.
References
1. Bennedsen, J.: Teaching and learning introductory programming – a model-based approach.
Ph.D. thesis, Oslo University, Oslo (2008)
2. Bergin, J.: Fourteen Pedagogical Patterns. http://csis.pace.edu/*bergin/PedPat1.3.html
3. Bergin, J., Roberts, J., Pattis, R., Stehlik, M.: Karel++: A Gentle Introduction to the Art of
Object-Oriented Programming. Wiley, New York (1996)
4. Bruner, J.S.: The Process of Education. Harvard University Press, Cambridge (1960)
5. Brusilovsky, P., Calabrese, E., Hvorecky, J., Kouchnirenko, A., Miller, P.: Mini-languages:
a way to learn programming principles. Educ. Inform. Tech. 2(1), 65–83 (1997). https://doi.
org/10.1023/A:1018636507883
6. Daly, T., Wrigley, E.: Learning Java Through Alice 3, 2nd edn. CreateSpace Independent
Publishing Platform, Charleston (2014)
7. Ehlert, A., Schulte, C.: Empirical comparison of objects-first and objects-later. In: 5th
International Workshop on Computing Education Research (ICER 2009), pp. 15–26. ACM,
New York (2009). https://doi.org/10.1145/1584322.1584326
8. Eremin, E.: Software system to learn objects. In: 5th Annual SIGCSE/SIGCUE ITiCSE
Conference on Innovation and Technology in Computer Science Education (ITiCSE 2000),
p. 188. ACM, New York (2000). https://doi.org/10.1145/343048.343212
9. Holland, S., Griffiths, R. Woodman, M.: Avoiding object misconceptions. In: Miller, J.E.
(ed.) 28th SIGCSE Technical Symposium on Computer Science Education (SIGCSE 1997),
pp. 131–134. ACM, New York (1997). https://doi.org/10.1145/268084.268132
10. Hubwieser, P.: Analysis of learning objectives in object oriented programming. In:
Mittermeir, R.T., Sysło, M.M. (eds.) ISSEP 2008. LNCS, vol. 5090, pp. 142–150. Springer,
Heidelberg (2008). https://doi.org/10.1007/978-3-540-69924-8_13
11. Kölling, M.: Introduction to Programming with Greenfoot Object-Oriented Programming in
Java with Games and Simulations, 2nd edn. Pearson, Boston (2016)
12. Kölling, M., Rosenberg, J.: An object-oriented program development environment for the
first programming course. In: Klee, K.J. (ed.) 27th SIGCSE Technical Symposium on
Computer Science Education (SIGCSE 1996), pp. 83–87. ACM, New York (1996) https://
doi.org/10.1145/236452.236514
13. Kölling, M., Rosenberg, J.: BlueJ – The Hitchhiker’s Guide to Object Orientation. Maersk
Mc-Kinney Moller Institute for Production Technology, University of Southern Denmark,
Technical report no. 2 (2002)
14. Lister, R., Berglund, A., Clear, T., Bergin, J., Garvin-Doxas, K., Hanks, B., Hitchner, L.,
Luxton-Reilly, A., Sanders, K., Schulte, C., Whalley, J.L.: Research perspectives on the
objects-early debate. In: Working Group Reports on ITiCSE on Innovation and Technology
in Computer Science Education (ITiCSE-WGR 2006), pp. 146–165. ACM, New York
(2006). https://doi.org/10.1145/1189215.1189183
15. Mannila, L., de Raadt, M.: An objective comparison of languages for teaching introductory
programming. In: 6th Baltic Sea Conference on Computing Education Research: Koli
Calling 2006 (Baltic Sea 2006), pp. 32–37. ACM, New York (2006). https://doi.org/10.
1145/1315803.1315811
16. Meyer, B.: Towards an object-oriented curriculum. In: Ege, R., Singh, M., Meyer, B. (eds.)
11th International Conference on Technology of Object-Oriented Languages and Systems
(TOOLS 1993), pp. 585–594. Prentice-Hall, Upper Saddle River (1993)
17. Pears, A., Seidman, S., Malmi, L., Mannila, L., Adams, E., Bennedsen, J., Devlin, M.,
Paterson, J.: A survey of literature on the teaching of introductory programming. In: Carter,
J., Amillo, J. (eds.) Working Group Reports on ITiCSE on Innovation and Technology in
Computer Science Education (ITiCSE-WGR 2007), pp. 204–223. ACM, New York (2007).
https://doi.org/10.1145/1345443.1345441
18. Pecinovský, R.: OOP – Learn Object Oriented Thinking and Programming. Eva & Tomáš
Bruckner Publishing, Řepín-Živonín (2013)
206 E.A. Eremin
19. Proulx, V.K., Raab, J., Rasala, R.: Objects from the beginning - with GUIs. In: 7th Annual
Conference on Innovation and Technology in Computer Science Education (ITiCSE 2002),
20. Ragonis, N., Ben-Ari, M.: On understanding the statics and dynamics of object-oriented
programs. In: 36th SIGCSE Technical Symposium on Computer Science Education
(SIGCSE 2005), pp. 226–230. ACM, New York (2005). https://doi.org/10.1145/1047344.
1047425
21. Ramalingam, V., Wiedenbeck, S.: An empirical study of novice program comprehension in
the imperative and object-oriented styles. In: Wiedenbeck, S., Scholtz, J. (eds.) 7th
Workshop on Empirical Studies of Programmers (ESP 1997), pp. 124–139. ACM, New
York (1997). https://doi.org/10.1145/266399.266411
22. Robins, A., Rountree, J., Rountree, N.: Learning and teaching programming: a review and
discussion. Comput. Sci. Educ. 13(2), 137–172 (2003)
23. Sajaniemi, J., Kuittinen, M.: From procedures to objects: a research agenda for the
psychology of object-oriented programming in education. Hum. Tech. 4(1), 75–91 (2008).
https://doi.org/10.17011/ht/urn.200804151354
24. Sanders, K., Boustedt, J., Eckerdal, A., McCartney, R., Moström, J.E., Thomas, L., Zander.
C.: Student understanding of object-oriented programming as expressed in concept maps. In:
39th SIGCSE Technical Symposium on Computer Science Education (SIGCSE 2008),
25. Sorva, J., Seppälä, O.: Research-based design of the first weeks of CS1. In: 14th Baltic Sea
Conference on Computing Education Research: Koli Calling 2014 (Baltic Sea 2014),
26. Tait, B.: Object orientation in educational software. Innov. Educ. Train. Int. 34(3), 167–173
(1997). https://doi.org/10.1080/1355800970340302
27. Tomek, I.: Microworlds for teaching concepts of object oriented programming. J. Univ.
Comput. Sci. 6(1), 423–434 (1995). https://doi.org/10.1007/978-3-642-80350-5_35
28. Uysal, M.P.: The effects of objects-first and objects-late methods on achievements of OOP
learners. J. Soft. Eng. Appl. 5(10), 816–822 (2012)
29. Xinogalos, S.: Object-oriented design and programming: an investigation of novices’
conceptions on objects and classes. Trans. Comput. Educ. 15(3), 1–21 (2015). Article 13.
https://doi.org/10.1145/2700519
30. Xinogalos, S., Satratzemi, M., Dagdilelis, V.: An introduction to object-oriented program-
ming with a didactic microworld: objectKarel. Comput. Educ. 47(2), 148–171 (2006).
https://doi.org/10.1016/j.compedu.2004.09.005
31. Zhu, H., Zhou, M.: Methodology first and language second: a way to teach object-oriented
programming. In: 18th Annual ACM SIGPLAN Conference on Object-Oriented Program-
ming, Systems, Languages, and Applications (OOPSLA 2003), pp. 140–147. ACM, New
York (2003). https://doi.org/10.1145/949344.949389
Real-Time Data Analyses in Secondary Schools
Using a Block-Based Programming Language
Andreas Grillenberger(B) and Ralf Romeike
Computing Education Research Group,

Friedrich-Alexander-Universität Erlangen-Nürnberg,
Martensstraße 3, 91058 Erlangen, Germany
{andreas.grillenberger,ralf.romeike}@fau.de
Abstract. Data management is central to many CS innovations: Smart

home technologies and the Internet of Things, for example, are based on
processing data with high velocity. One of the most interesting topics
emphasizing several challenges in this field is real-time data analysis. In
secondary CS education, it is only considered marginally. So far, there are
no tools suitable for general-purpose real-time data analysis in school. In
this paper, we discuss this topic from a secondary CS education perspec-
tive. Besides central concepts and differences to traditional data analysis
using relational databases, we describe the development of a general-
purpose Snap! extension that allows accessing and processing data from
various sources. Thereby, students are enabled to conduct data analy-
ses using, for example, sensor data or web APIs. With the example of
a weather station, we outline how this tool can be used in school for
analyzing sensor data generated in the classroom.
Keywords: Real-time data analysis · Data stream systems · Data man-

agement · Sensor data · Physical computing · Secondary CS education
1 Introduction
In previous years, several innovative topics of computer science became perva-
sive in our daily lives. The increasing possibilities when capturing, processing,
analyzing and visualizing data are, for example, central to home automation
(“smart home”) and when connecting various devices in the Internet of Things.
Also, data analyses are often used for decision-finding, even in contexts in which
this would hardly be expected in, such as autonomous cars or when addressing
voters in election campaigns. The emerging field data management comprises
several developments originating from the challenge to store and analyze so-
called big data, i.e. large amounts of data that are generated and analyzed with
high velocity and have strongly varying structures [10]. Despite its highly inno-
vative character, various practices and principles of this field seem promising
for general educative CS education. For example, partitioning and replication of
data are increasingly relevant, not only when developing data management sys-
tems, but also when synchronizing data between different devices and services:
https://doi.org/10.1007/978-3-319-71483-7_17
208 A. Grillenberger and R. Romeike
Understanding why data are sometimes duplicated or lost and how to prevent
this is hardly possible without getting to know the underlying concepts, such as
redundancy. One exemplary topic emphasizing several principles and practices of
data management, is real-time data analysis. In practice, such analyses are often
conducted using data stream systems. With these, immediate analysis of large
amounts of data are possible. As today we are often confronted with the results
and implications of real-time data analyses, getting to know their basic princi-
ples becomes increasingly relevant for understanding such results, for evaluating
their quality and relevance, but also for making decisions based on such data
[7]. For developing and fostering competencies related to capturing, storing and
processing data and for estimating and evaluating the consequences and implica-
tions of these possibilities, real-time analyses are an ideal starting point. Hence,
in this paper we will outline the functionality and central principles of real-time
data analysis in general and of data stream systems in particular. Based on this,
we will describe how central ideas of real-time data analysis can be included in
secondary education using a general-purpose data stream system extension for
the block-based programming language Snap! [8].
2 Related Work
Despite their high relevance, not only in computer science, neither the topic
data management in general, nor real-time data analyses in particular, have
been examined in detail as topics for secondary CS education yet. Although
related contexts, such as the Internet of Things, have already been discussed as
topic for CS education (cf. e.g. [11]), courses and projects presented are typically
designed for higher education. As they pursue different goals, only few aspects
can be transferred to our work, such as the practical orientation and the hands-
on approach. Also in robotics, processing streams of sensor data and events is
central from a technical perspective, but has not yet been discussed as a topic
for CS education. As we have shown in a qualitative analysis of various curricula
and educational standards, today most data management topics, including real-
time data analysis, are typically not included in secondary CS teaching [5]. In
consequence, as of today, there are no suitable tools for general-purpose real-time
data analysis in school. To bring aspects of modern data analyses to school, we
already developed and described a tool for analyzing the Twitter data stream
[6], which has been presented to teachers and discussed with them at various
opportunities. The advantages in comparison to data analyses using databases
were convincing for most teachers. Yet, there was always one concern: To work
with the Twitter API, students need to have an account on this platform. As a
workaround, we offered the possibility to work offline using a cached data set, but
this was obviously less motivating for the students than working with live data.
Also, teachers were often concerned that privacy issues might arise and noted
the restricted flexibility of the tool, as it only supported the Twitter data stream
as a data source. Hence, for tackling these issues, in this paper we describe a new
Real-Time Data Analyses in Secondary Schools 209
general-purpose approach which allows using various data sources and nearly all
possibilities of the continuous query language1 .
3 Real-Time Data Analyses
Today, real-time analyses are being used for several purposes: analyzing credit
card transactions in order to prevent fraud (cf. [12]), reacting to temperature
changes in smart home environments, or monitoring the environment of smart
cars. Although in many use cases, providing results of data analysis immediately
is not essential, in most modern use cases there are at least weak restrictions
on the reaction times of a system: Typically, users want data to be available
as soon as possible, hence “soft real-time”2 becomes increasingly important. In
other use cases, such as industrial robots, the restrictions on data analysis are
even harder: they need to fulfill firm or hard real-time requirements3 .
In order to meet a deadline even when analyzing large amounts of data, tra-
ditional data analyses are not sufficient. Instead, typical real-time data analy-
ses are in several parts completely different. While traditional data sources are
rather finite and discrete, for example when sensor data streams are analyzed,
the analysis system needs to handle infinite and continuous data sets (cf. [13]).
Also, the amount of data sources (in particular sensors) is growing, with wireless
communication the data are available faster than years ago, and the data rate is
increasing drastically [13]. Yet, when regarding current CS teaching, analyzing
data in schools, if at all, typically takes place using databases: For example, in
popular weather station projects, the sensors data are often stored in relational
databases and queried using the query language SQL. While this approach is
suitable for traditional data analyses, several challenges occur when trying to
conduct real-time analysis that way: In particular when data are generated con-
tinuously, analysis jobs would need to be started frequently in order to ensure
up-to-date results, as determining distinct points in time for starting the analy-
sis is not possible when analyzing data streams4 . In consequence, the database
(and the analysis) gets overloaded or stuck because of the continuous and paral-
lel read and write operations. Additionally, using databases for this purpose also
results in enormous amounts of data being stored, because of the high rate of
1
The “continuous query language” (CQL) is similar in syntax to SQL, but in particu-
lar allows using “sliding windows”, which makes it suitable for data stream analysis
(cf. [1]).
2
Soft real-time allow even frequent misses of the deadline, as only service quality is
being influenced (cf. e.g. [3]).
3
Hard real-time strongly requires adherence to a deadline, as exceeding it results in a
system failure. Firm real-time tolerates missing the given deadline infrequently, but
the analysis results become irrelevant after the deadline and the quality of service is
degraded.
4
“A data stream is a real-time, continuous, ordered (implicitly by arrival time or
explicitly by timestamp) sequence of items. It is impossible to control the order in
which items arrive, nor is it feasible to locally store a stream in its entirety.” [4].
data generation and as it is not possible to delete outdated values, as information

would get lost. Hence, although databases are a suitable tool for storing large
amounts of rather static data for a long time and analyzing such data, they can
hardly be used for real-time data analysis. Today, this problem can be solved by
using data stream systems, which are designed as general-purpose systems for
fast analysis of continuous data streams.
In several characteristics, these systems are the opposite of databases: In par-
ticular, data stream systems do not store data permanently, but instead process
them on-the-fly by applying queries that have been previously defined [9]. Thus,
data stream systems are appropriate when data are to be analyzed only once,
immediately after being generated. This is the case in particular for sensor data:
As every value is only up-to-date until the next value is generated, immediately
analyzing them is important. In addition, in such use cases it is typically rea-
sonable to drop values if they cannot be analyzed immediately and to continue
with the next value which is then the most relevant one, as significant changes
typically do not influence single values only. Data stream systems can, in partic-
ular, show their potential when “sliding windows” are used, i.e. when a defined
number of recent values or all values from a defined time span are analyzed
(cf. [2]). With databases, during every execution of a query, all data stored in
the database would be re-evaluated (whether or not they were changed before),
while using “sliding windows” allows to cache and evaluate only the relevant
data, ensuring that they are deleted from the cache when out-of-date.
Data
Data
Stored
Data
Queries
Queries Results
Results
Fig. 1. Functional principles of databases and data stream systems.
The different functional principles of databases and data stream systems are
visualized in Fig. 1: When using databases, storing and analyzing data are inde-
pendent processes. In contrast, in data stream systems, data is directly analyzed
after they were received by the system. Hence, many modern systems process-
ing continuous data streams are based on characteristics of data stream systems
rather than of databases: For example, when measuring the wind velocity in a
smart home project for triggering the closure of windows, it is neither important
to process every single value nor to store these data permanently. Hence, data
stream systems are ideal for this purpose. To summarize, Table 1 shows a com-
parison of central characteristics of both, databases and data stream systems.
Table 1. Comparison of the main characteristics of databases and data stream systems.
Characteristics Database system Data stream system

Data storage Permanent storage, often No storage except caching of
relational but also other values needed for further
paradigms processing
Data processing When executing a query Immediately after new data
were received
Typical data Data that should be stored Continuous data streams
permanently and that are consisting of data which
relevant for long term; often quickly become outdated;
used for multiple typically only processed for
queries/purposes; often one purpose
rather static data
Query language Typically descriptive, often Typically descriptive, often
SQL CQL
Popular Professional Research implementations:
implementations implementations: MySQL, STREAM, AURORA,
PostgreSQL, MongoDB . . . TelegraphCQ, Odysseus . . .
Central concepts Integrity, consistency, Availability, concurrency
redundancy
4 Real-Time Data Analyses in Current Secondary CS

Teaching
When the topic “data” is addressed in secondary CS teaching, the focus is typ-
ically on (relational) databases [5]. This has not been changed since the early
1990s. Thus, because of the tremendous developments in recent years, several
topics that are strongly related to the students’ daily life today are only consid-
ered marginally, such as data security, privacy, encryption or meta-data. Also,
“real-time data analysis” is typically not considered in secondary CS teaching at
all: Although such analyses are pervasive in our daily life, understanding their
underlying principles and functioning is not a goal of typical CS education. Yet,
to understand their relevance, opportunities and threats, getting to know basic
aspects of real-time data analysis is essential. For example, it is hard to imagine
how large amounts of data captured by CCTV may be analyzed in real-time
without knowing the underlying analysis methods. But when they know about
the restrictions and differences of real-time analysis in comparison to data analy-
sis in general, students also become familiar with the involved opportunities and
threats.
Recently, aspects of data management and real-time data analysis can be
found in school teaching when data is acquired and evaluated in physical com-
puting projects. For example, sensors are used for measuring and evaluating
environmental influences. In simple projects, data is typically processed directly
on the microcontroller, e.g. when just reacting on concrete values or measuring

and displaying values such as the current temperature. But in more complex
projects, for example calculating average temperatures for defined time frames,
analyzing the data directly on the microcontroller is often not possible due to
memory and performance restrictions. Hence, data is often sent to a computer
for further storage and analysis. This typically involves multiple tools, such as
the programming environment for the microcontroller, one for controlling data
storage and analysis, and a database such as SQLite for data storage.
Currently, a popular project in secondary schools is building weather sta-
tions. With several sensors, data is gathered continuously. Often, the goal is
more complex than just showing the current values, as for example, average or
extreme values are more interesting. When calculating such values for defined
time frames (i.e. for the last 24 h), the analysis becomes too complex for typical
microcontrollers used in school. When trying to use traditional approaches for
storing and analyzing the sensor data, all values in this time frame need to be
stored in a suitable data storage, deleted from it as soon as they are outdated
and replaced with other values. This would lead to numerous read/write opera-
tions. When using databases for this purpose, as it is common in such projects,
the problems described before arise. Yet, professional data stream systems are
hardly suitable for use within CS education because of their high complexity.
5 Challenges for Teaching Real-Time Data Analysis

5.1 Data Sources
As analyzing data in real-time is an especially suitable for dynamically changing

data, data sources that are traditionally used in CS education are not suitable
for this purpose. Instead, there are in particular two approaches for accessing
“live data”:
Web APIs. With several web applications providing free access to their applica-
tion programming interfaces (APIs), lots of data can be acquired. For example,
social media platforms such as Twitter and Facebook provide access to large
parts of their data. There are two approaches for accessing the data: While
Twitter uses a “push” method, i.e. the client subscribes to the stream and is
notified by the server about new tweets (comparable to the observer pattern),
most other APIs use a “pull” approach, i.e. the client frequently requests infor-
mation. Although pulling produces more unnecessary communication and delays,
it has an advantage for teaching: As the connections are not kept open all the
time, typically multiple users can use the same credentials. Also, as most APIs
are based on the REST principle and hence are accessed via HTTP calls, using
them is relatively easy, particularly if they are based on a “pull” approach. Hence,
these interfaces are ideal data sources for real-time data analysis, as the data are
(depending on the application) highly dynamic and as access is typically only
limited by rate limits of the APIs.
Sensor Data. Another data source is sensor data. With physical computing
projects gaining popularity in CS education, using sensors for measuring data
from the environment of a system has already become common in CS teach-
ing. This is also a suitable approach for gathering data for data management
lessons: The main advantage of using sensor data as a basis for data analy-
sis is that it is generated in real-time in the classroom environment, without
hurdles of using web APIs, such as mandatory user accounts or privacy issues.
Also, as nearly all common programming languages allow communication with
microcontrollers, from traditional object-oriented languages such as Java right
up to block-based programming environments designed for educational use such
as Scratch or Snap!, finding a suitable tool for obtaining data is not a problem.
Another advantage of using sensor data in data management education is the
shift of focus from just processing and analyzing data, to the whole data life
cycle from their acquisition and modeling, through processing and analysis, to
visualization and (perhaps) deletion of data. This depicts the real usage of data
and shows how different CS fields come together in innovative topics like the
“Internet of Things”, which is based particularly on data management technolo-
gies and (interconnected) microcontrollers. As capturing sensor data can be done
in various ways, there are also very different types of sensors that might be used,
and hence also different ideas for data processing and analysis.
5.2 Development of a Real-Time Data Analysis Extension for Snap!
Design Decisions. For accessing both, sensor data and REST APIs, several
tools suitable for CS education already exist. In particular, the block-based pro-
gramming language Snap! [8] has high potential for data management teaching:
Not only does it provide blocks for sending HTTP requests that are suitable
for accessing REST APIs, it cam also communicate with Arduino (and compat-
ible) microcontroller boards in the fork Snap4Arduino5 . Hence, it is possible to
access sensor data captured with such boards. It is also highly expandable, so
that processing data is possible using the same tool with which they are cap-
tured, and it is easy-to-use for students, even without previous experiences using
it. Thus, for conducting real-time data analysis in CS teaching, we implemented
the central features of data stream systems and CQL [1] in the Snap! program-
ming environment. Using Snap! as a basis clearly contributed to our main goals:
The tool should be flexible enough to be used with various data sources, it should
be easy-to-use in a school context and it should be easy-to-extend by teachers
and (ideally) also by the students themselves. To allow the data stream system
extension to be used in Snap! and all its forks, we only used functionalities that
Snap! provides in its end-user interface and avoided to directly modify its source
code. Hence, the extensions is completely based on primitive Snap! blocks and
JavaScript functions (which can also be called in Snap! out-of-the-box by map-
ping them into blocks). In addition to extending Snap! for allowing data stream
analyses, we also extended it for data visualization purposes in the same way.
5
http://snap4arduino.org.
Implementation of the Data Stream Analysis Extension. In contrast to

our tool for analyzing the Twitter data, which was based on imperative queries,
for the Snap! data stream system extension (“Snap!DSS”) we use declarative
queries. This approach facilitates handling the queries as one unit in the pro-
gramming environment and allows nesting queries like is possible, for example, in
typical query languages such as SQL. Also, as professional data stream systems
typically use declarative languages such as CQL, Snap!DSS conveys the function
of professional data stream analysis tools better than SnapTwitter. For making
the extension as flexible as possible, Snap!DSS offers the following functionalities:
– Creating a new data stream from any data source (“reporter block” in Snap!).
– Combining data from several sources in one stream.
– Running queries on data streams, using aggregate functions, projections,
selections and sliding.
– Using analysis results as a data source for a new data stream (nesting).
– Continuous evaluation of queries in the background.
To implement these functions, we created a data structure which is repre-

sented in Snap! as an unevaluated block. This internal data structure is rather
complex: The data stream system, data streams and queries are represented as
lists with various fields (cf. Fig. 2). In particular, a data stream stores a list of
its queries and information on its data source; each query stores its parameters
as well as a cache of current values. When a data stream system is started, each
query of all of its streams is executed in the background about once per second6 .
Hence, the values stored in the data stream system are updated permanently,
as it is common in data stream systems. Despite its complex structure in the
background, Snap!DSS can be easily used with the following blocks:
– : Creates a new data stream system, to which streams

can be added for continuous processing.
– : Creates a new data stream, which is based on the
data given in an input field.
– : Adds a data stream to a data stream system.
– : Executes queries on the data streams.
After the first use of a query, the system continuously captures the relevant
data for the query while running. In a query, aggregate functions (average,
minimum, maximum, sum) and windows may be used.
– : Starts continuous data processing in the background.
– : Stops background data processing.
Besides allowing high flexibility and all relevant functionalities, our approach
also gives the teacher the opportunity to simply adapt the tool for concrete
lessons and thus to further reduce complexity (e.g. by hiding blocks or by creating
new ones), without losing the flexibility for complex analyses.
6
We limited processing of queries to one time per second to prevent performance
issues. Yet, this limit can easily be changed by modifying the block in Snap!.
attribute
names
function
attributes
in query
window size
current
queries
cached values
unit of window
data sources
id of date
attribute
Fig. 2. Internal representation of a data stream in Snap!.
Implementation of the Data Visualization Extension. To enable students

to visualize data they captured and processed using the data stream system, we
also extended Snap! in order to allow graph visualizations. Therefore, we use
the JavaScript library plotly.js 7 , which is used in offers diverse plot styles. For
using this library in Snap!, we use the JavaScript block to load and initialize
the API directly in Snap!. This implementation is relatively easy, shows how
Snap! may be extended that way and is prototypical for including JavaScript
libraries in Snap!. In our extension, the plotly.js library is loaded on-the-fly in
the background, thus the user does not have to cope with loading or initializing
it and can use it like any other block. Currently, the functions shown in Fig. 3
are implemented: The C-shaped “plot” block prepares the “plotly.js” library and
(after defining the traces to plot) generates the image shown on the Snap! stage.
Within the “plot” block, users can define multiple traces as well as horizontal
lines (which are, for example, useful for indicating average values).
Fig. 3. Blocks for creating data visualizations using plotly.js.

7
https://plot.ly/javascript/.
6 Example Project: Weather Station

As mentioned before, a common example when capturing and processing sensor
data in school, is building a weather station. This project allows students to
capture lots of data with sensors using almost any typical physical computing
platform8 . In order to avoid the problems of using the typical database-based
approach, in the following we will describe, how a weather station can be imple-
mented using our Snap!DSS extension.
In this project, students can recognize the main functional principles and
concepts of real-time data analysis. Hence, they can carry out their own data
analysis and understand how easily several data sources can be combined for
achieving better analysis results, e.g. by making recorded values more accessi-
ble through aggregation and by creating simple forecasts (e.g. using pressure
differences as an indicator for weather changes). Also, by including API data,
the results could be compared to professional analysis for evaluating them. For
capturing and analyzing the data, Snap4Arduino on the software side (with
Snap!DSS loaded) and an Arduino Uno microcontroller board on the hardware
side can be used. We use a combined sensor for temperature, pressure and humid-
ity, which we connected using the “grove shield”9 . Yet, our extension does not
restrict using other sensors, as long as their values can be read using reporter
blocks in Snap!, as those are needed for using the data stream extension. Using
our extension, the values read by the reporter blocks are interpreted as a data
stream. After adding this stream to the data stream system and starting it, the
sensor values are evaluated according to the queries defined. As no queries have
been defined yet, nothing would happen when running the program; first, at
least one query has to be defined (which can even happen on-the-fly while the
system is running). As we are interested in the minimum, average and maximum
values of both, pressure and temperature, we can define three queries, as it is
depicted in Figs. 4 and 5 for calculating the average value of a light sensor.
The results of the queries can directly be shown, e.g. by clicking on a query
block, but they can also be used as input for other blocks in Snap!. Hence, we
are, for example, able to show them on a LCD display or to visualize them
using the data visualization extension, as it was done in the example shown in
Fig. 5. For displaying the data of one sensor and the corresponding average value,
the code and the result is shown in Fig. 6. The visualizations may be changed
without losing information, because the data from all sensors are analyzed in
the background as defined in the queries after starting the system.
This example project (Fig. 7) is easier to understand and implement than
the typical implementation in schools using databases, in particular as fewer
different systems are involved. Due to the design of the tool, it is clearer and
allows to be extended flexibly, and it can also be used for more complex analy-
ses. The project can be realized with students with or without prior knowledge
8
An exemplary project was described by the Raspberry Pi Foundation: https://www.
raspberrypi.org/blog/school-weather-station-project/.
9
http://wiki.seeed.cc/Grove Starter Kit v3/.
Fig. 4. Analyzing a sensor data stream.
Fig. 6. The line chart represents the

actual values, the horizontal line shows
the average value.
Fig. 5. Visualizing sensor data.
on data management or physical computing, and with no additional material or

systems in comparison to typical physical computing projects. The concept can
be transferred to other projects, such as building a smoke detector (which could
also take into account temperature increases, in order to prevent false alerts),
realizing smart home projects (e.g. switching on the light when the environmen-
tal brightness becomes lower than the average in the last five minutes, ensuring
that temporary peaks are neglected), or even for quantified self projects (such
as realizing a sleep quality measurement device or for measuring heart rates).
Fig. 7. Exemplary weather station project using a combined temperature and pressure
sensor as well as a LCD attached to a grove shield.
7 Summary
In this paper, we discussed central characteristics of real-time data analyses.
Based on these, we proposed a tool, which is suitable for conducting general-
purpose data stream analysis in secondary CS education. In an example, we
have shown how this tool can be used in school in combination with physical
computing. Although the Snap!DSS extension reduces complexity, it preserves
characteristics and functionalities of real-time data stream analyses. It is oriented
at the syntax and semantics of CQL, allows its typical operations and implements
central aspects and functionalities of general-purpose data stream systems. In
combination with accessible hardware, students are enabled to pursue their own
ideas and to conduct analyses on their own, which are otherwise only possible for
professionals in the field. Thus, this approach opens up new opportunities for the
students to benefit from the innovations in data management. Yet, by preserving
the typical character of data stream systems and real-time data analyses, it also
fosters a better understanding of several central concepts of data management.
References
1. Arasu, A., Babu, S., Widom, J.: The CQL continuous query language: semantic
foundations and query execution. VLDB J. 15(2), 121–142 (2006)
2. Babcock, B., Babu, S., Datar, M., Motwani, R., Widom, J.: Models and issues
in data stream systems. In: Proceedings of the 21th ACM SIGMOD-SIGACT-
SIGART Symposium on Principles of Database Systems, pp. 1–16. ACM, New
York (2002)
3. Lee, E.A., Seshia, S.A.: Introduction to Embedded Systems. MIT Press Ltd., Cam-
bridge (2017)
4. Golab, L., Özsu, M.T.: Processing sliding window multi-joins in continuous queries
over data streams. In: Proceedings of the 29th International Conference on Very
Large Data Bases (VLDB 2003), vol. 29, pp. 500–511. VLDB Endowment (2003)
5. Grillenberger, A., Romeike, R.: A Comparison of the field data management and
its representation in secondary CS curricula. In: Proceedings of WiPSCE 2014.
ACM, Berlin (2014)
6. Grillenberger, A., Romeike, R.: Analyzing the Twitter data stream using the
snap! Learning environment. In: Brodnik, A., Vahrenhold, J. (eds.) ISSEP 2015.
978-3-319-25396-1 14
7. Grillenberger, A., Romeike, R.: Teaching data management: key competencies and
opportunities. In: Brinda, T., Reynolds, N., Romeike, R. (eds.) Proceedings of
KEYCIT 2014. Universitätsverlag Potsdam (2014)
8. Harvey, B., Mönig, J.: Snap! Reference manual (2014). http://snap.berkeley.edu/
SnapManual.pdf. Accessed 09 July 2017
9. Krämer, J.: Continuous Queries over Data Streams - Semantics and Implementa-
tion. Philipps-Universität Marburg (2007)
10. Laney, D.: 3D data management: controlling data volume, velocity, and variety.
Technical report, META Group, February 2001
11. Mäenpää, H., Varjonen, S., Hellas, A., Tarkoma, S., Männistö, T.: Assessing IoT
projects in university education: a framework for problem-based learning. In: Pro-
ceedings of the 39th International Conference on Software Engineering, pp. 37–46.
IEEE Press, Piscataway (2017)
12. Mayer-Schönberger, V., Cukier, K.: Big Data: A Revolution That Will Transform
How We Live, Work, and Think. Houghton Mifflin Harcourt (2013)
13. Nittel, S.: Real-time sensor data streams. SIGSPATIAL Spec. 7(2), 22–28 (2015)
XLogoOnline: A Single-Page, Browser-Based
Programming Environment for Schools Aiming
at Reducing Cognitive Load on Pupils
Juraj Hromkovič1 , Giovanni Serafini1 , and Jacqueline Staub1,2(B)

1
Department of Computer Science, ETH Zürich,
Universitätstrasse 6, 8092 Zürich, Switzerland
2
Pädagogische Hochschule Graubünden, Scalärstrasse 17, 7000 Chur, Switzerland
{juraj.hromkovic,giovanni.serafini,jacqueline.staub}@inf.ethz.ch
Abstract. For more than twelve years, our chair has been introduc-
ing primary school children to algorithmic thinking by teaching them
how to program in Logo. The key element of the proposed didactic app-
roach consists in reducing the extraneous cognitive load on the pupils.
We developed and stepwise refined the required teaching materials that
allow for introducing only a few instructions in a programming language,
which is gradually extended simply relying on modular design. XLogoOn-
line is our new browser-based, single-page programming environment for
schools which is perfectly attuned to our curriculum. We argue that the
platform reduces the extraneous cognitive load on the pupils thanks to a
heavily-simplified workflow, appropriate for young children, and present
evaluations that confirm high usability and acceptance across ages.
1 Introduction
It is crucial that young pupils leave school not only as passive users of com-
puters but also with the ability to think algorithmically and solve problems
by programming. This form of learning is constructive, enriches creativity and
teaches precision but is strenuous and demanding by its very nature. In this
work, we reflect on the difficulties children encounter while learning to program
and introduce our two-pronged approach of a programming environment and
curriculum as technical and didactic solutions to reduce cognitive load on novice
programmers.
1.1 At Long Last, Computer Science in Swiss Schools

Research and development of programming concepts and languages have a long
and rich academic tradition in Switzerland. Pascal, Modula-2, and later Oberon
were conceived by Swiss computer science pioneer and Turing-award recipi-
ent Niklaus Wirth, with his main objectives being that university students are
introduced to programming using modern and didactically appropriate teaching
material.
https://doi.org/10.1007/978-3-319-71483-7_18
220 J. Hromkovič et al.
It may seem paradoxical, however, that Switzerland is still debating about

the opportunity to introduce computer science as a mandatory subject at pri-
mary, lower and higher secondary school. While the German speaking part of the
country slowly seems to agree on computer science for all children up to 15 years,
the so-called “Plan d’étude Romand” in the French part of the confederation still
completely neglects computer science [10]. At present, higher secondary school
students all around the country are allowed to choose computer science as an
optional subject, usually in the last year of the curriculum. The Swiss Conference
of Cantonal Ministers of Education is finally considering to introduce Computer
Science as a compulsory subject at higher secondary school. For the ongoing
debates a decision is expected for October 2017.
For more than twelve years, our chair at ETH Zurich – together with several
regional partners, among them Pädagogische Hochschule Graubünden [9] and
University of Basel – have been conducting projects introducing primary school
children and their teachers to programming in Logo [3–5,11] using our didactic
approach. In the last two years, in the greater area of Basel alone, 96 primary
school classes (1862 pupils) and their teachers were taught how to program using
our teaching materials. Currently, we are even active outside the Swiss borders
and are introducing 8 primary school classes (141 pupils and their teachers) to
programming in the Principality of Liechtenstein.
In the rest of this paper we present XLogoOnline, our new browser-based pro-
gramming environment for introducing primary school children to programming.
1.2 Organization of the Paper
In Sect. 2, we reflect on the contribution of programming on algorithmic thinking

with a focus on preserving pupils’ working memory. In Sects. 3 and 4, we highlight
technical and didactic design principles of our curriculum before honing in on
implementation details of XLogoOnline. Section 5 introduces objectives, setup
and results of three experiments after which we finally summarize the key insights
of the paper in Sect. 6.
2 Algorithmic Thinking and Cognitive Load
We consider the expressions algorithmic thinking and computational thinking as

two equivalent and interchangeable ways to state the same concept [5]. Knuth
already reflected 30 years before Wing about the relations between algorithmic
thinking and mathematical thinking, and informally described algorithmic think-
ing as his “perception of the most typical thought processes used by a computer
scientist”. Aho captured the nature of algorithmic thinking as follows: “We con-
sider computational thinking to be the thought process involved in formulating
problems so their solutions can be represented as computational steps and algo-
rithms” [7]. Irrespective of whether we refer to it in one way or another, we believe
XLogoOnline: A Single-Page, Browser-Based Programming Environment 221
that computer science should be introduced at school “as an independent sci-

entific subject” and that it should be “studied both for its intrinsic intellectual
and educational value and for its applications to other disciplines” [2].
2.1 Programming is a Key to Algorithmic Thinking
Following the definition of Aho, we are firmly convinced that learning how
to develop well-structured programs is a key element to mastering algorithmic
thinking, and an ideal way to experience scientific methods at an appropriate
level of abstraction. In our curriculum, pupils first learn to write small pieces
of code, which then are tested and revised for improvement. We mainly focus
on programming-in-the-small and the curriculum trains pupils to implicitly for-
mulate a hypothesis (write a program to solve a given problem), verify their
hypothesis (run and test the program), and successively refine the hypothesis as
necessary.
Modular design is both a technique to develop well-structured programs and
an overall problem solving strategy that can be used to tackle any complex tasks,
even outside of computer science. In our programming classes, pupils are explic-
itly taught to seek geometric figures for repetitive patterns. They rapidly discover
the advantages of programming and naming such small bite-size elements, and
of reusing them in a proper modular fashion. The pupils learn to successively
decompose problems into smaller tasks which are each solved independently, and
finally reassembled to solve the original assignment.
Learning to program is generally considered to be a complex cognitive activ-
ity. In our didactic approach, programming-in-the-small and modular design
require the pupils to organize their programs into conceptually independent
units. This modularization process is core to teaching computer science: First, it
helps to approach challenging problems by breaking them down into smaller
tasks, which can be solved independently by applying top-down or bottom-
up problem solving strategies. Second, thinking about problem instances in an
abstract way helps identifying structural properties inherent to the problem,
and therefore allows the pupils to classify them into abstract problem classes,
with common properties that can be solved relying on one single parameterized
program. This results in a high intrinsic cognitive load. Educational research
highlights, that this kind of cognitive load is specific to the contents and consid-
ers it as immutable [14].
2.2 How We Reduce the Extraneous Cognitive Load
The educational materials we developed consist of the German textbook

Einführung in die Programmierung mit Logo [6], the booklet Programming in
LOGO [1], and Logo programming environments that are tailored for young
pupils and programming exercises with a visual feedback [1]. The main subject
of our paper is XLogoOnline, the most recent programming environment in this
suite [12,13].
We strongly believe that Logo – despite its age – still is one of the most suit-
able languages for introducing novices of all ages to programming. Moreover, we
are convinced, that Logo together with our teaching materials and the program-
ming environments allow for a noticeable reduction of extraneous cognitive load.
Regarding knowledge dimensions presented in the Revision of Bloom’s Taxon-
omy [8], we argue that our proposed didactic setting lower the mental effort and
preserve the working memory of the pupils:
1. We massively reduce the impact of factual knowledge: Instead of introducing
pupils to all the instructions and features of a specific programming language,
we deliberately focus on a very limited set of basic instructions and teach the
kids how to gradually extend the programming language by naming their pro-
grams and making them available for reuse. Furthermore, the chosen syntax
is adequate to age and abstraction skills of school children. The pupils are
not merely using a programming language but they are developing their own,
personalized vocabulary to actively interact with the computer.
2. Conceptual knowledge is introduced in a compact and didactically effective
way: in the eyes of the pupils, a program simply is a sequence of unambigu-
ous instructions that the computer understands and predictably executes.
Furthermore, we focus on only one looping control structure that avoids vari-
ables, and, later on, we introduce the simpler notion of a static parameter,
as a preparation step to the challenging concept of variable. Variables are
postponed to lower secondary school.
3. The procedural knowledge and to some extent also metacognitive knowledge
developed by children mainly relies on modular design. We use this design
principle as a straightforward problem solving strategy and, moreover, as a
mean to hide nested loops.
Readers who are interested in further details and reflections on our teaching
approach should refer to our prior publications [3–5,11]. With the exception of
the textbook, all teaching materials and environments are freely available to all.
2.3 Programming Languages for School Education

The choice of a specific programming language for school education strictly
depends on the targeted educational objectives. We mainly aim at designing
(and not only at using) a language together with the children, and strongly
believe that programming classes are a unique chance to learn being precise in
a semantic but also in a syntactic way. A text-based programming language
such as Logo is inherently well-suited for teaching both precision and modular
design. Block-based programming languages assume that the syntax is a major
hurdle while learning how to program, and therefore try to lower the extraneous
cognitive load by using graphical elements and a drag-and-drop programming
approach. Although the general objectives of programming classes with block-
and text-based languages seem to be similar, the methods adopted for reducing
the cognitive load on the pupils considerably differ and make separate research
on this topic necessary.
3 Technical and Didactic Design Principles

XLogoOnline was designed as a browser application to meet the requirements of
being an easily deployable programming environment which may be used across
platforms. At schools we encounter a plethora of old and modern computer
infrastructures that vary from stationary computers to laptops and tablets, that
have different operating systems, various screen sizes, and all may or may not
be connected to the Internet.
We had to decide upon numerous technical and didactic design alternatives.
In the following, we present the most influential choices we have taken and
describe their impact on the implementation of our platform:
1. Built-in diagnostics that offer smart hints and pinpoint errors are automati-
cally generated in XLogoOnline and play an important role in the interaction
between computer and pupil. They help to foster a mostly autonomous and
self-driven learning process, and moreover alleviate pressure on the teacher.
2. XLogoOnline adopts a two-pronged strategy to reduce distractions in the
user interface. First, the chosen single-page approach, which makes all infor-
mation for developing and understanding a program visible at one glance,
prevents obstructively layered windows. Second, exhibiting nothing but the
most important features makes for a simple and clear user interface.
3. Our reactive, browser-based, client-side application relies on state-of-the-art
technologies and runs on a local web-server in offline mode. It uses the local
file system to store and retrieve code from earlier sessions. This enables the
platform to be deployed independent of whether an internet connection is
available or not.
4. Our programming environment relies on Canvas, a native dot-matrix data
structure for drawing supported by most modern browsers, which does not
pollute the Document Object Model and, moreover, which provides a handle
for conveniently downloading pictures. Our tool makes efficient use of the
canvas by discerning whether flushing the entire drawing to the canvas is
possible, or whether it should be split into sequences (e.g. in case of a wait
command) as the parser is running.
4 Implementation of XLogoOnline
In the following, we explain what goes on behind the scenes when a Logo instruc-
tion is executed by touching upon the parser, the editor and the canvas and
explaining what actions are triggered by a simple press of the <enter> button
in the input field.
4.1 Information Flow

Pupils interact with four main components (Fig. 1): they issue commands to the
input field in a line-by-line manner and each line contains either built-in or user-
defined commands. These programs are defined in the code editor and retrieved
from it during execution. Once execution has finished successfully, the result is
visible as a drawing in the rendering component and the command is added to
the history which keeps track of all recently issued commands. Next, we zoom
in on the pertinent implementation aspects of each component.
Fig. 1. Information flow
4.2 Editor
While our curriculum only introduces a very basic set of Logo primitives, pupils
use the editor to extend the set of commands by defining new customized pro-
grams. Programming demands precision and mistakes are inevitable, therefore
we provide helpful suggestions in terms of diagnostics which highlight token
classes directly in the editor by using different colors and inline markers to pin-
point syntactical errors. Programs are typically declared in the following char-
acteristic form:
TO PROGRAMNAME :param1 :param2 ... :paramN
body
END
Two especially common mistakes we see in pupils’ programs are due to mis-
taken declarations or invocations of programs. The former usually appear as
parameters with missing double-colons, which are interpreted as invocations
to non-existing programs. The latter often manifest as missing spaces between
program names and their arguments, leading to a single program without any
parameters. Calling such a program later causes a run-time error because the
according signature cannot be found. We help identify these classes of mistakes
using syntax highlighting and checking, which we will explain in the following.
Syntax Highlighting. We assign unique colors to each class of syntactic ele-

ments and reflect those in the editor. For instance, built-in commands like fd
(for forward) or cs (for clear screen) are rendered in a different shade than
numbers, parameters and mathematical constants. Over time, pupils become

acquainted to this coloring scheme and common cases of invalid input such as
missing spaces or double colons, become evident. This simple, yet surprisingly
effective mechanism helps pupils resolve basic mistakes autonomously and fur-
ther enhances their understanding of grammatical concepts by revealing how the
code is interpreted.
Syntax Checking. As the users type in the editor, their programs are contin-
uously and automatically interpreted in the background after a short period of
inactivity. Programs which fail to parse are analyzed further to locate missing
keywords and erroneous numbers of parameters. The exact location and type of
an error is signaled by adding a red cross directly beside the faulty code which
shows a hint when hovered over. Doing so aids pupils and frees them from con-
centrating on syntactical details and let them focus on semantics instead. Next,
we discuss the underlying concepts of our interpretation process.
4.3 Parser and Interpreter

The parser starts by validating that programs satisfy the structural requirements
imposed by the language specification. Using a grammar, the sequence of char-
acters found in the program is transformed into a sequence of tokens by the
lexer. Next, a parser derives a parse tree from the tokens. We label these two
processes as parsing. The grammar is defined hierarchically as a set of recur-
sive rules. XLogoOnline parses programs starting at the top-level rule and this
starting rule differs in two cases: in the input field only program invocations
are allowed; in the editor, however, the starting rule only permits program dec-
larations. Once parsed, a decision is made and ill-typed programs are rejected
and otherwise execution continues with an interpretation phase which traverses
the parse tree using a visitor. Each node corresponds to a Logo instruction and
execution entails drawing to the canvas. These two steps we label drawing.
4.4 Canvas
Once a program has successfully executed, the result is shown on a drawing

surface which allows the user to drag with their mouse and explore the painting.
To support panning, we draw onto a large backing buffer consisting of the visible
view port in the middle and its eight neighboring cells. The view port can fill at
most the whole screen, leading to a backing buffer nine times the actual screen
size. The view port consists of two canvases; one used to display the turtle, the
other to draw strokes onto. Strokes are not rendered to the display until explicitly
flushed. This is a costly operation because the whole frame needs to be redrawn
and so, where suitable, we limit these operations and defer the flush to the end
of the execution.
5 Classroom Studies and Performance Evaluation

We evaluated XLogoOnline based on three experiments. First, we measured the
cognitive load experienced by pupils and teachers across several age ranges while
programming in Logo. Next, we compared XLogoOnline against one of its pre-
decessors in terms of usability. Last, we benchmarked three programs in a per-
formance sanity test measuring elapsed time for drawing and compilation.
5.1 Programming and Cognitive Load Across Age Groups
In a typical classroom, a single teacher will guide and assist the entire class;
pupils learning how to program, however, progress at differing rates. Ideally
pupils should work autonomously while not experiencing cognitive overload. In
order to observe children programming in a real classroom environment, we ran
several introductory lessons, measured pupils’ and teachers’ exhaustion levels
using a questionnaire and examined the differences in cognitive load across ages.
Setup and Participants. We introduced 75 children (23 girls and 52 boys)

and 12 teachers to programming in Logo. In groups of 12, the (mostly) novice
programmers spent 75 min solving exercises using sequences of basic movement
commands. Each group consisted either of (i) children aged 6 to 9 being exposed
to geometry for the first time, (ii) children aged 10 to 13 more used to text-style
exercises, geometry and abstract thinking, or (iii) teachers who have progressed
even further in their education but mostly lack prior knowledge in computer
science. Our hypothesis is that with increasing age students become more accus-
tomed to abstract thinking and their cognitive load decreases. Cognitive load,
however, cannot be measured in absolute terms; so we instead asked partici-
pants about their level of exhaustion by having them fill out a questionnaire
after programming. Answers were provided along a 4-option Likert-type scale
featuring smiley faces. This subjective, indirect measure was proposed in prior
work by Paas et al. [15], which assumed that the felt complexity of a task and
the cognitive load are correlated.
Results. The combined student data confirmed that children indeed feel cogni-
tively exhausted while programming (Fig. 2a) reflecting the need for a tool which
eases their learning whilst also not imposing too much extraneous cognitive load.
The more detailed breakdown by age in Fig. 2b confirms our hypothesis; the
shifting peaks reflect that teachers are least exhausted, followed in turn by the
older and the younger pupils respectively who showed increased exhaustion.
Since exhaustion levels proved to be similar and rather high in both student
groups, we require an objective measure which helps differentiating more finely.
For this reason, we plan to run a follow-up study examining cognitive load by
measuring the number of exercises solved in total. Furthermore, we will explore
the influence of different exercise types on the cognitive load of the programmer.
Fig. 2. Questionnaire indicate particularly high cognitive load for young pupils.
The experiment confirmed that most children experience rather high cog-
nitive load while programming, which reinforces the need for tailored teaching
material to enable better learning experiences. Children aged 6–9 showed the
highest exhaustion levels, while children aged 10–13 were not only less exhausted,
but we also observed them to be more progressive and able to work without direct
guidance.
5.2 Usability Comparison of XLogo and XLogoOnline
To understand whether our design choices had a positive impact on user expe-
rience, we ran a second user study and examined the children’s interaction with
XLogoOnline, their understanding of the purpose of selected components therein
and general impression of the user interface. Figure 3 shows a side-by-side com-
parison of what solving the same exercise would look like in XLogoOnline and
its predecessor XLogo.
Fig. 3. XLogoOnline features an uncluttered single-page layout with bigger fonts and
a reduced set of GUI features for reducing the extraneous cognitive load and in support
of a young audience whose fine motor skills have yet to fully develop.
Setup and Participants. A fifth-grade class consisting of 18 pupils, gener-

ally aged 11 or 12 years, participated in this on-site experiment. The class was
first introduced to the traditional programming environment XLogo for three
weeks where they learned about the basic commands as well as the concepts of
repetition and modularization in Logo.
After getting used to programming in XLogo for three programming sessions
which lasted two hours each, we introduced XLogoOnline while covering regular
polygons. After a short introduction, we left the children to solve exercises for two
hours before finishing the experiment by handing out feedback questionnaires
covering three main questions: first, we checked their understanding of the use
of the four main UI components by letting pupils draw connections between
pictures of both user interfaces and descriptions of the components’ purpose;
second, we asked pupils to grade the four components in XLogoOnline on both
their visual appearance and their functionality; last, we asked them to tell us
about their general impression working with the platform.
Results. A vast majority (83% of all pupils) understood the usage and mean-
ing of the four UI components correctly and could easily transfer from XLogo
to XLogoOnline. We received strongly positive feedback (praises for the syn-
tax checker, error pinpointing, layout, clarity and comprehensibility) which give
reassurance that simplifying the user interface in XLogoOnline, adopting a flat
design and equipping it with additional diagnostic tools like syntax highlighting
and syntax checking are indeed beneficial. Some pupils missed certain features
XLogo provided and others found the space usage not optimal, which we sub-
sequently adjusted according to their needs. We conclude, that the transition
between XLogo and XLogoOnline posed no significant burden on the students.
5.3 Performance Test
It is easy to build complex shapes involving several tens of thousands of drawing

instructions by simply repeating sequences of basic patterns many times over. To
ensure that even these complex exercises in our teaching materials are executed
in reasonable time, we chose three benchmark programs (shown in Fig. 4) which
represent a range from short to more time-intense programs and checked their
performance.
Fig. 4. Benchmark programs ranging from cheap through expensive

The respective programs were re-run 1000 times in XLogoOnline while mea-
suring their overall execution times. We found an average performance of 20000
drawing instructions per seconds on a four-year old off-the-shelf laptop. Conse-
quently, the running time for simple shapes like the ones shown in Fig. 4a and
b add up to merely a few milliseconds. Moreover, even the most time-intense
exercises in our curriculum run to completion on the order of seconds, which we
consider satisfactory for the typical classroom setting.
Finally, we investigated the breakdown of execution time between drawing
and parsing to determine the cost of diagnostics, which rely on repeated pars-
ing of the input. We found drawing to be the more influential factor between
the two phases and that, as programs get bigger, it increasingly dominates the
overall execution time (shown in Fig. 5). Diagnostics like error classification and
additional hints are potentially expensive, however, the time spent in parsing
constitutes a negligible fraction of the total time and significantly helps pupils
recover from their mistakes.
Fig. 5. The larger the program, the more time we spent traversing the parse tree (i.e.
drawing) while building the parse tree (i.e. parsing) takes only a fraction of the time.
In summary, we found XLogoOnline able to execute all exercises in our teach-

ing material in reasonable times and parsing to be relatively cheap which allows
for further diagnostics as an additional help for the pupil, supporting a more
autonomous diagnosis and remedy of syntactic errors.
6 Conclusion
Teaching how to program in hundreds of schools, we found algorithmic thinking
and programming to be exciting but exhaustive activities, especially for young
pupils. We introduce XLogoOnline, our new browser-based programming envi-
ronment that aims at reducing the extraneous cognitive load on school children.
It is tailored around a curriculum with a deliberately small set of basic com-
mands. Pupils steadily learn to expand the language with custom commands
while solving algorithmic problems. As a distinguishing attribute, our program-
ming environment features diagnostics which provide useful feedback to the pro-
grammer while having an uncluttered and easy to use single-page user interface.
The programming IDE can be deployed at the click of a button using only a web
browser and, thanks to its reactive design, can be run on a broad set of devices
with different screen sizes and platforms. We ran classroom studies investigating
cognitive load during introductory lessons and found a clear need for appropriate
didactic settings which reduce the load on pupils’ working memory. XLogoOnline
was shown to be intuitive and generally approved by primary school pupils.
Future Work: Our broader goals are to understand what troubles young pro-
grammers the most. Web environments are especially well-suited for collecting
rich data sets, for instance, containing student mistakes. Analyzing this data
may reveal patterns which can be tied back to conceptual error classes and tack-
led more explicitly in our curriculum. Further, we are investigating social effects
in programming and have built an early prototype extension for XLogoOnline
that allows pupils to collaborate on exercises by sharing code and drawings with
their classmates.
References
1. Logo-Uunterrichtsmaterialialien & -Programmierumgebungen. http://www.abz.
inf.ethz.ch/logo. Accessed 18 Sept 2017
2. Gander, W., et al.: Informatics education: Europe cannot afford to miss the boat.
Informatics Europe & ACM Europe Working Group on Informatics Education
(2013)
3. Hromkovič, J., Kohn, T., Komm, D., Serafini, G.: Combining the power of Python
with the simplicity of logo for a sustainable computer science education. In: Brod-
nik, A., Tort, F. (eds.) ISSEP 2016. LNCS, vol. 9973, pp. 155–166. Springer, Cham
(2016). https://doi.org/10.1007/978-3-319-46747-4 13
4. Hromkovič, J., Kohn, T., Komm, D., Serafini, G.: Examples of algorithmic thinking
in programming education. Olympiads Inform. 10, 111–124 (2016)
5. Hromkovic, J., Kohn, T., Komm, D., Serafini, G.: Algorithmic thinking from the
start. Bull. EATCS 121, 132–139 (2017)
6. Hromkovič, J.: Einführung in die Programmierung mit LOGO, 3rd edn. Springer,
Wiesbaden (2014). https://doi.org/10.1007/978-3-8348-2266-6
7. Knuth, D.E.: Algorithmic thinking and mathematical thinking. Am. Math. Mon.
92(3), 170–181 (1985)
8. Krathwohl, D.R.: A revision of Bloom’s taxonomy: an overview. Theor. Pract.
41(4), 212–218 (2002)
9. Matter, B.: Projekt “Programmieren in der Primarschule”. http://www.abz.inf.
ethz.ch/wp-content/uploads/2015/03/InfobroschAug2010-3.pdf. Accessed 18 Sept
2017
10. Parriaux, G., Pellet, J.-P.: Computer science in the eyes of its teachers in
french-speaking Switzerland. In: Brodnik, A., Tort, F. (eds.) ISSEP 2016.
978-3-319-46747-4 15
11. Serafini, G.: Teaching programming at primary schools: visions, experiences, and
long-term research prospects. In: Kalaš, I., Mittermeir, R.T. (eds.) ISSEP 2011.
LNCS, vol. 7013, pp. 143–154. Springer, Heidelberg (2011). https://doi.org/10.
1007/978-3-642-24722-4 13
12. Staub, J.: XLogoOnline. http://ethz.ch/xlogo. Accessed 18 Sept 2017
13. Staub, J.: Xlogo online - a web-based programming IDE for Logo. Master Thesis
(2016). https://e-collection.library.ethz.ch/view/eth:49742?lang=en
14. Sweller, J.: Cognitive load theory. Psychol. Learn. Motiv. 55, 37–76 (2011). Aca-
demic Press
15. Paas, F.G.W.C., van Merriënboer, J.J.G., Adam, J.J.: Measurement of cognitive
load in instructional research. Percept. Motor Skills 79(1), 419–430 (1994)
Introduction to Bebras Challenge Management:
Overview and Analyses of Developed Systems
Valentina Dagiene, Gabriele Stupuriene(B) , Lina Vinikiene,

and Rimantas Zakauskas
Vilnius University Institute of Mathematics and Informatics,

Akademijos Street 4, 08663 Vilnius, Lithuania
{valentina.dagiene,gabriele.stupuriene,lina.vinikiene}@mii.vu.lt
rimantas.zakauskas@gmail.com
Abstract. The paper analyses contest management systems used by the

multicultural Bebras challenge community. Starting as a single annual
contest in 2004 in a few countries, the Bebras challenge on Informat-
ics and Computational Thinking has spread to over 60 countries and
successfully developed a worldwide network. A requisite tool to perform
the challenge is a Contest Management System (CMS). Countries have
been developing various systems, depending on their practice, the num-
ber of participants, financial issues, etc. The overview and analyses of
the Bebras CMS used by various countries are presented here. The main
focus concerns the Lithuanian Bebras CMS, its development, architec-
ture, functionalities, and thoughts for future improvements.
Keywords: Bebras challenge · Computational thinking · Contest

management system · CMS · Informatics education
1 Introduction
The Bebras challenge1 on informatics and computational thinking (CT) is an
educational community network which brings together teachers and scientists
who are interested in teaching informatics at primary and secondary schools.
The aim of the Bebras challenge is to engage students in informatics, deepen
their understanding in computer technologies, and promote learning by solving
short informatics concept-based tasks [1]. The tasks are a crucial point of the
challenge: they should involve at least one informatics concept, arranged with
an interesting story, be attractive, short, and interactive or dynamic [2].
In general, the challenge is divided into six age categories: Pre–Primary
(grades 1–2), Little Beavers (grades 3–4), Benjamin (grades 5–6), Cadet (grades
7–8), Junior (grades 9–10), and Senior (grades 11–13). The core part of the
Bebras challenge is a contest organized by each country in the second and
third weeks of November. The Southern Hemisphere countries (Australia, New
Zealand, and Malaysia) have a different time schedule. Additional rounds of the
1
http://bebras.org.
https://doi.org/10.1007/978-3-319-71483-7_19
Introduction to Bebras Challenge Management 233
contest and other problem-solving activities are organized in various countries.

In Lithuania, the Bebras challenge is organized in two rounds: a school-wide
contest in November and a nation-wide competition for selected high school stu-
dents in January of the following year. During the school-wide round students
perform tasks using computers and are supervised by teachers in their schools.
The second round of the challenge is supervised by universities or colleges in var-
ious regions. In Lithuania, over 33 thousand participants performed tasks in the
first round (2016) and 385 of them were invited to the second round in January
2017.
A contest management system (CMS) is an essential environment for man-
aging the challenge effectively and efficiently. It should support simple tools,
which enable task development, users’ management, grading, announcement
areas, records of solutions, reports, and data storage. In order to implement
the Bebras challenge, several contest management systems have been developed
and adapted to users and their needs (described in Sect. 3).
In this paper, we focus mainly on the system that supports the Lithuanian
Bebras challenge on informatics and computational thinking. The paper is orga-
nized as follows: Sect. 2 describes some examples of CMSs used in various coun-
tries. In Sect. 3 we review a survey that was provided to the Bebras community.
Main aim of this survey was to gather information about management and imple-
mentation of the Bebras challenge, as well as about the systems used, and to
get an understanding of the differences of the basic challenge management prin-
ciples. Section 4 presents a more detailed description of the Lithuanian Bebras
contest management system, with the improvements from its first deployment
until today. We end with a Conclusion.
2 Examples of the Contemporary Contest Management

Systems
The International Olympiad in Informatics2 (IOI) is one of the most popular
competition in Computer science. Students work individually on a set of Com-
puter Science assignments. At the IOI the use of a grading system (contest
management system) is an absolute necessity. This system provides all the nec-
essary content for the students to successfully prepare and participate in the
programming contests, as well as all the necessary capabilities for the organi-
zation of programming contests, including submission of tasks and automatic
grading [3]. The CMS (used for IOI 2012) was organized in a modular way, with
different services, running potentially, on different machines. When applicable,
services can be replicated to allow scaling for larger contests [4]. Whatever the
system offers, it should be designed such that it is simple, robust, secure and
flexible [5].
The Bebras challenge is not a programming contest, so requirements for this
system are different. A short presentation of the existing Bebras systems was
published in the paper [6].
2
http://www.ioinformatics.org.
234 V. Dagiene et al.
A Finnish CMS3 was built using Rails, a rapid web development frame-
work using the programming language Ruby. In time, features were added to
the system by different developers. Some of these features break the separation
between models, views, and controllers. For example, although the system was
designed to be database-agnostic, its current version occasionally accesses the
SQL database directly and is therefore strongly tied to M ySQL.
The CMS of France4 is one of the most optimal and well-developed: in this
system, an optimized front-end is used which reduces the number of requests
from clients to the server and consequently the load on a server. While the web
workload is distributed, all the web servers access a single relational database.
The sessions are implemented using M emcached technology, reducing the load
on a database while still maintaining a single session across all the web servers.
If the web servers become unavailable, competitors are provided with a coded
message at the end of their competition. They can send this coded message to
the organizers using e-mail and have their results entered into the system. To
minimize the communication between the web server and the clients, results
are only submitted at the end of the competition, with no backup in case a
competitor web browser crashes.
In 2013, Slovenia started to develop a new Bebras CMS5 that have to be
[6]: high performance, scalable, and fault-tolerant. To achieve that, a three-layer
architecture consisting of a front-end layer, a business logic layer, and a distrib-
uted database back-end layer was applied [7]. As shown in Fig. 1, the front-end
was implemented using HT M L, CSS and JavaScript, the business logic layer
was implemented in P HP using the Y ii web development framework, while
M ySQL Cluster was used as the database back-end.
Fig. 1. The architecture of the Slovenian Bebras CMS [6]
3
http://www.majava-kilpailu.fi.
4
http://concours.castor-informatique.fr/.
5
http://bober.acm.si/.
The Bebras challenge in Russia and Belarus is supported by distance contests

execution/educational system. The system is implemented using Java, P lay
f ramework, M ongoDB, and can support different types of competitions, such
as the contest, “Construct, Test, Explore”. All tasks are uploaded to the system
by filling fields, such as a statement, question, a set of answers or interaction,
and an explanation containing the correct answer, “It’s Informatics”. Two types
of tasks are used:
– Multiple-choice questions with the opportunity to select one of four variants.

– Dynamic tasks. These tasks were implemented in JavaScript. There are
libraries that allow the creation of drag-and-drop, text input field tasks.
Both types of questions provide the opportunity to select “I don’t know”. Par-
ticipants in the challenge, school organizers who manage participants from their
schools, higher-level organizers (who manage the school organizer), etc. are roles
supported by the system [8].
Pikl et al. characterize the ideas for the competition systems and websites,
such as logical layout, important information visible at all times, frequently asked
questions, discussion between visitors, and a dynamic web page [9]. The history,
results, information about the competition, who the competition is for, contact
form and additional information should be provided at the competition site.
3 Survey of the Bebras CMS in Various Countries
In this section we describe the process of collecting the data and the results
from analysing these data. A questionnaire with fourteen open questions was
announced in May of 2016 and was accessible until the end of February (2017).
The aim of this questionnaire was to collect information about CMSs in differ-
ent Bebras community countries, to understand the real situation about CMSs
(what is common, what is different), and to discover valuable suggestions for
others. Thirty-two countries answered the questionnaires regarding the Bebras
challenge. Table 1 shows the distribution of countries by responsibilities of CMS
support (organizers in countries themselves create and develop the system; sup-
port of the system is trusted for the private company6 ; organizers use platforms
available on the Internet).
Sweden and Finland collaborated in the development of the same system.
Their system is implemented using Ruby and interactive tasks are created using
JavaScript. Serbia uses Slovenia’s well-developed system (from 2013). Turkey
uses LMS M oodle (however, they do not have interactive tasks). This is easy
to use, although the interface should be changed and there should be a special
template for the contest [10]. Croatia also uses LMS M oodle. Teachers there
take care of editing, publishing tasks, and managing participants with M oodle,
since every student has their own access to the system. Belarus uses the
6
https://www.eljakim.nl/project/beverwedstrijd/.
Table 1. Countries distribution by responsibilities of CMS support
CMS developed by Belgium, Bulgaria, Cyprus, Estonia, Finland, France,

organizers Hungary, Indonesia, Italy, Latvia, Lithuania, Macedonia,
Russia, Slovakia, Slovenia, Spain, Taiwan, Ukraine
CMS developed by a Canada, Germany, Ireland, Japan, The Netherlands,
company Romania, Singapore, Switzerland, USA
P latf orms Croatia (Moodle), Turkey (Moodle), Belarus (Yandex
Contest)
Y andex.Contest system7 . It is similar to the ACM-like contest system where

it is possible to check test-like questions. Macedonia prefers to use a web-based
system, developed with node.js. The main features of this system are: multiple
browsers and devices support, statistics collection, data backup, and it can also
sustain the loss of a server or database.
The crucial point for CMSs is the number of people who participate at the
same time. It partially depends on the settings of the system. However, small
countries with fewer participants do not measure this. For example, for France
(more than 470000 participants) and for Germany (more than 290000 partici-
pants) the maximum number of participants is 10 000. The French platform is
designed to handle much more, if needed. For Belarus, Croatia, Lithuania, and
Turkey, the maximum number of participants at the same time is 1000, and for
Bulgaria, Finland, and Italy, it is 500 participants. Slovenia is able to connect
the largest number of participants at the same time (more than 20,000). There is
no limitation for contestants in Ukraine because they work offline, their answers
are recorded as files, and are collected afterwards.
In general, the Bebras challenge is held online and participants solve tasks
using computers, but in some countries students have the opportunity to
solve tasks using other devices. For example, in France, Germany, Italy, The
Netherlands, Macedonia, Spain, and USA students can solve tasks using any
computer with a browser, an Android, or iOS tablets. In Indonesia, Japan and
Latvia students use desktop computers, laptops, tablets, and mobile phones. In
Belarus, Canada, Estonia, Hungary, Serbia and Singapore only computers are
used. In some cases, only computers are recommended due to interactive tasks,
because tasks may not work correctly, or the site is not optimized for mobile
phones. In Slovenia, from second grade to fifth grade, tasks are completed on
paper and from sixth grade, computers are used.
Using CMS options or additional analytics tools, organizers can provide data
about their system, devices, technical details, the number of participants, and
answer statistics for those participants. For example, France and Russia col-
lect information about devices or browser versions through Google Analytics.
Belarus has installed Y andex.M etrica. Twenty of thirty-two countries gather
information about gender and seven countries are not interested in the gender
7
https://contest.yandex.ru.
issue. In some cases, such as Italy and Singapore, the possibility of indicating
gender is optional. All countries collect the number of participants by age group,
but in some CMSs, such as in Belarus, France, and Romania, participants are
listed with their precise ages. Personal information, such as name, surname,
school, or language is presented in individual cases. It depends on the country’s
attitude to privacy rules and data publication. Some privacy rules forbid using
the student data for statistical research.
The questionnaire confirms that the Lithuanian Bebras CMS has the most
developed data collection and statistics analysis. Nevertheless, sixteen countries
indicated the need to gather data about gender, devices used, etc., in each coun-
try. Discussions were also recorded about a privacy policy, depending on what
kind of data were recorded.
Bebras tasks can be of any type, from very basic multiple-choice tasks, to
interactive tasks where students manipulate objects visually, or even to advanced
interactive tasks where students can write and run small programs. Interactivity
is very typical for computers; thus, it is clear that a computer-oriented contest
should apply interactive elements to explain or solve tasks. These interactive
tasks are often “funny” and easy to understand. However, the interactive ele-
ments require a lot of effort by the programmer to implement them. According to
the answers from the questionnaire, twelve countries use only multiple choice or
open-ended questions for the Bebras tasks. Nineteen countries provide and inter-
active tasks during the Bebras challenge. Eighth countries out of the nineteen
surveyed countries, that use interactive tasks, use the Bebras Lodge tool. This
is a special tool for creating and developing interactive (dynamic) tasks8 . France
has developed its own special Bebras CMS, which manages all interactive Bebras
tasks during the challenge. They use JavaScript with RaphaelJS library, as well
as their own set of libraries 9 . Other technologies used for interactive task imple-
mentation are HT M L5 (Italy) and JavaScript (Estonia, Finland, Japan, and
Russia). Seven countries (out of thirteen, which do not have interactive tasks)
plan to introduce interactive tasks in their CMS. Nevertheless, interactive tasks
installation depends on their systems’ facilities.
Experience with contest management inspires those interested to think about
new system features or improvements for the present CMS. French Bebras orga-
nizers would like to have a tool for teachers for creating interactive tasks (some-
thing similar to the Bebras Lodge tool, but with a different approach). Germany
would like to have an API to import, store, and export Bebras tasks including
their complete interactivity (the Bebras Pool). Lithuania has a plan to collect
data regarding how many times participants have a second look at the same task
or return to resolve that task, and how many times they have changed the answer.
Slovakia wants to measure the time spent on each task. Ukraine would like to do
compatibility with mobile devices and developer-friendly animations. Belarus,
Canada, Croatia, Indonesia, Serbia and Turkey are planning to add different
types of interactive tasks or develop more interactive task features. Macedonia
8
http://bebras.licejus.lt/login.
9
https://github.com/France-ioi/bebras-modules/tree/master/pemFioi.
would like to add more functionality to its CMS to create a friendlier environ-
ment for the teacher (for example, see information about students). Macedonia
also emphasized the importance of testing as ensuring the performance of par-
ticipants. Singapore is planning to introduce reports on results for schools and
students into their CMS.
4 Bebras CMS Development: Lithuanian Experience

At the beginning of the Bebras challenge in Lithuania (2004), we used PDF
technology. Organizers were faced with the basic problem of software installa-
tion, students’ answers upload, and delivery to the central administration. The
Lithuanian Bebras CMS was developed in 2010 and named as the Bebras chal-
lenge field10 . The Lithuanian Bebras CMS was is as a tool to organize the contest,
manage information about participants (students) and teachers, to gather data
on task solutions, to manage the challenge, and provide detailed statistics and
reports. The system was tested by exploitation (more than 30 000 student’s
entered the system during last year) and efficiency-designed for managing chal-
lenge. More than 5 500 new accounts were created for primary and secondary
school students last year. The number of new user accounts is growing by similar
additional accounts each year.
4.1 System Description

The Lithuanian Bebras CMS is dedicated to create, manage, and administer the
challenge online. It works as a web application. The system is based on three-
level architecture (similar to the CMS of Slovenia) (Fig. 2). CMS works as a web
application and consists of a set of subsystems, which has a well-design interface:
– System administration (security, back up, resource monitoring, etc.),
– User management (registration system, authentication, user profile),
– Challenge management (creation, administration and monitoring),
– School administration (official list of all schools in Lithuania). A school list
is updated in accordance with cooperation with the Centre of Information
technologies in Education every year.
– Tasks management (create, import tasks),
– Results and communication management (participants and teachers can dis-
cuss about particular tasks, preview statistical data).
The Lithuanian Bebras CMS is developed by using the M ySQL relational data-
base management system (DBMS), Apache HT T P server and P HP program-
ming language. The operational system for web server is Linux. CMS uses
the Model-View-Controller structural pattern, which means that an application
should be separated from its presentation into three main parts: the model, the
view, and the controller. In Model-View-Controller, the view displays informa-
tion to the user and together with the controller comprises the application’s user
10
http://lt.bebras.lt.
Fig. 2. The architecture of the Lithuanian Bebras CMS
interface [11]. A relational database is created to store details of tasks, students

and teachers. The data are stored in different tables and relations are established
using primary keys or other keys known as foreign keys. CMS is built to be com-
patible with all operating systems and the latest versions of browsers, although
the use of Internet explorer or Edge browsers are not recommended.
During the week of the challenge various statistical data are collected in
order to gather information about students’ abilities to solve tasks. In addition,
this information helps to improve informatics education and improve teacher
professional development.
4.2 Functional Requirements
In this section we discuss the Bebras CMS functionality using Use case diagram
(Fig. 3). System administrators have full access, including management of task:
creation and importation from the Bebras Lodge tool. Teachers are provided
with challenge access to their school’s students and have access to the results of
their students. Teachers register their students and system administrators enroll
them to the system (it helps to avoid cheating). Registered teachers can con-
firm students’ participation in the challenge during the school-wide challenge
in November. Furthermore, teachers have the opportunity to preview the tasks,
participate in the discussion, and print certificates for their own students. Stu-
dents have access to the challenge during the Bebras week and can preview the
tasks, and comment on or discuss the particular tasks after the challenge week.
Students can see their results only after completing the challenge.
Fig. 3. Use case diagram of the Lithuanian Bebras CMS
CMS design is modular (consists of 11 modules) according to its functional-

ity. Users management module includes actions such as create, edit, confirm,
delete user, preview or edit data, preview the list of participants, archive par-
ticipants, give a new password, move student to a higher class, search user,
send news and reminders for the participant, give permission for the teacher to
coordinate the challenge, and generate the reports for students’ solutions.
Challenge management module involves these actions: create and edit
the task set, set the challenge date and time, preview task in the task set, task
testing, manage permission to solve task set, participation in the challenge, and
answer survey about particular tasks.
System administration module consists of previewing the list of schools,
editing participants’ and teachers’ data, searching (by school title, teacher
name/surname, student name/surname), and creation of survey for participants.
Tasks management module is designed to create, edit, copy, upload,
delete, preview, export task, filter task by tags, create tags for the tasks, and
search tasks by name or ID. System supports multiple-choice questions with the
opportunity to select one correct answer from four. Other questions, such as ani-
mated tasks, text input field tasks, and drag and drop tasks are imported from
the Bebras Lodge tool. For multiple-choice tasks, the administrator fills these
fields: task ID, title, task description and possible answers, answer comment if
it is needed, chosen task difficulty, age group, and language.
Results preview module provides the opportunity to revise the desired

student, class solutions. Tasks are shown with the marked correct/incorrect
answer, students’ choice, and point. Challenge results for registered participants
are stored and teachers can view them.
Statistical and report module is used to review data about the time
taken to solve the tasks, answers, and the number of participants. We collect the
following data about participants: name, surname, gender, grade, and school.
Also, we gather data about the type of devices and browsers the participants
use in the challenge; how much time spent when solving the task for the first
time; how much time in seconds the participant spends on every task (the sum
of seconds is counted if the participant returns to solve the task). Each solution
is separated by the student ID number generated by the system, therefore all
statistics can be compared. In addition, data about the students’ numbers in
Lithuanian schools are saved in the system as well, so we can compare how
many students in a particular school participated.
Front-page. Module includes login to the system, a preview of participants’
results according to the municipality, age group, and a report about participants’
numbers in Lithuania. Training is available on the front-page without any special
registration. The time of training is limited to sixty minutes and does not depend
on the student’s age. There is no limit to the number of answer submissions.
Training results for unregistered users are displayed immediately after finishing
the session.
Authentication. Module implements the user registration, login and log-out
function, and password reminder. Participants are able to login with a F acebook
account. When a user registers with the system for the first time, the system
automatically sends an email with the link, and the user has to approve it. An
email is a required field on the system’s registration form for purposes of authen-
tication and personal data. Primary school pupils are faced with the problem of
not having a personal email account. For this reason, their teachers can upload
the list of participants to the system. The system generates passwords and user
names automatically from the uploaded file.
Certificate module. The system automatically generates the filled certificate
that is prepared by an administrator after the challenge is completed. Teach-
ers get certificates for organizing and coordinating the challenge. Students who
participated in the first or second round of the challenge receive certificates for
participation. About 10% of the best students from each age group get win-
ners’ diplomas. These diplomas are printed and students receive them during
the awards ceremony.
Language module enables localization and adapts the system for a spe-
cific language by translating resources. Using discussion module users can
follow discussions (read, comment, delete, edit). As helpdesk email, phone, and
F acebook are used for communication purposes.
The special system feature is the management of two challenge rounds. The
differences between the first and second round is the permission to solve tasks
managed by the system administrator. During the first round, the local organizer
(teacher) is responsible for the students and marks students who can solve tasks.
It can be done only during the scheduled time. Before the second round, the
system administrator has to mark the students who are involved in the challenge.
Students get email with the confirmation link and students, who confirm their
email, solve tasks during the second round. In addition, task sets of the first and
second round are separated, therefore the system administrator can easily follow
the history of a task set (when and which tasks were used in the challenge). The
Bebras CMS is used as the questions pool. About 2000 tasks are collected in the
system.
4.3 Practicalities
The Lithuanian Bebras CMS is used as a tool for contest management, but
sometimes we demonstrate the system for teachers or systems developers to
provide them with practice, understanding the challenge policy, and teacher
training. Unfortunately, the Lithuanian system does not provide teachers with
the possibility of preparing their own challenge to use in the educational process.
If they want to practice tasks development, we recommend a Bebras Lodge tool.
Each year a systems developer adds new functionalities to the system or
improves existing components. For example, there was a new opportunity in
2016 to upload the participants’ list. Next year we are planning to introduce
the badge policy as a participation engagement tool. Students will be able to
earn badges depending on the collected points. We hope to have four types of
badges: gold, silver, bronze medals, and a certificate of participation. Badges do
not always affect learners’ motivation as an inspiration to reach better results
or be involved in the learning process, but we hope it will work as an award
that all students will receive. The second addition to the system will be a two-
dimensional system, which will enable the classification of tasks according to
computational thinking and informatics concepts. Detailed analysis of the new
Bebras tasks classification is provided by [12].
5 Conclusions
This paper discusses CMSs as a tool to manage the Bebras challenge in various
countries. Answers from thirty-two (out of a total of 39) countries running the
Bebras challenge were collected and analysed. The results show that 18 counties
developed their own CMSs. 12 countries used only multiple choice or open-ended
answers for the Bebras tasks, and 18 countries provided interactive tasks. In the
future, some countries would like to use more interactive task features, as well
as develop compatibility with mobile devices. The Lithuanian Bebras CMS was
developed in 2010. The system consists of eleven modules, and has a three-layer
architecture. One of the future plans is integration a two-dimensional system
for enabling the classification of tasks according to computational thinking and
informatics concepts.
Acknowledgement. We acknowledge the international Bebras Community as well as

Centre of IT Development of Vilnius University for support. Thanks Carol Sperry, our
longtime colleague at informatics education, for proofreading the paper and making
suggestions.
References
1. Benaya, T., Zur, E., Dagiene, V., Stupuriene, G.: Computer science high school
curriculum in Israel and Lithuania - comparison and teachers’ views. Baltic J.
Modern Comput. 5(2), 164–182 (2017)
2. Dagiene, V., Stupuriene, G.: Bebras-a sustainable community building model for
the concept based learning of informatics and computational thinking. Inf. Educ.
15(1), 25–44 (2016)
3. Kostadinov, B., Jovanov, M., Stankov, E.: A new design of a system for contest
management and grading in informatics competitions. In: ICT Innovations 2010,
Web Proceedings, pp. 87–96 (2010)
4. Maggiolo, S., Mascellani, G.: Introducing CMS: a contest management system.
Olympiads Inf. 6, 86–99 (2012)
5. Jovanov, M., Kostadinov, B., Stankov, E., Mihova, M., Gusev, M.: State compe-
titions in informatics and the supporting online learning and contest management
system with collaboration and personalization features MENDO. Olympiads Inf.
7, 42–54 (2013)
6. Kristan, N., Gostiša, D., Fele-Žorž, G., Brodnik, A.: A high-availability bebras com-
petition system. In: Gülbahar, Y., Karataş, E. (eds.) ISSEP 2014. LNCS, vol. 8730,
pp. 78–87. Springer, Cham (2014). https://doi.org/10.1007/978-3-319-09958-3 8
7. Gostisa, D.: Doctoral dissertation: Visoko razpolozljiv tekmovalni sistem v oblaku
za tekmovanje Bober-CaaS, Univerza v Ljubljani (2014)
8. Pozdniakov, S.N., Kirynovich, I.F., Posov, I.A.: Contest “Bebras” on informatics
in Russia and Belarus. Olympiads Inf. 10(Special Issue), 55–65 (2016)
9. Pikl, B., Kristan, N., Ham, Z., Brodnik, A.: Developing integrated ACM competi-
tions website. In: International Electrotechnical and Computer Science Conference
(2012)
10. Kalelioğlu, F., Gülbahar, Y., Madran, O.: A Snapshot of the first implementation of
Bebras international informatics contest in Turkey. In: Brodnik, A., Vahrenhold, J.
(eds.) ISSEP 2015. LNCS, vol. 9378, pp. 131–140. Springer, Cham (2015). https://
doi.org/10.1007/978-3-319-25396-1 12
11. Leff, A., Rayfield, J.T.: Web-application development using the
model/view/controller design pattern. In: Proceedings of the 5th IEEE Interna-
tional Conference on Enterprise Distributed Object Computing (EDOC 2001),
pp. 118–127. IEEE Computer Society, Washington (2001)
12. Dagiene, V., Sentence, S., Stupuriene, G.: Developing a two-dimensional cate-
(2017)
Author Index
Aarne, Onni 15 Mayerová, Karolína 91

Monga, Mattia 102
Barendsen, Erik 65, 155 Morpurgo, Anna 102
Brondino, Margherita 169
Budinská, Lucia 91
Nordén, Lars-Åke 117
Calcagni, Annalisa 102
Ohashi, Yutaro 129
Dagiene, Valentina 141, 232
De Gradi, Debora 169
Parnes, Peter 117
du Plessis, Susanna H. 77
Pasini, Margherita 169
Pears, Arnold 141
Eremin, Evgeny A. 195 Peltola, Petrus 15
Pieterse, Vreda 77
Grgurina, Nataša 65 Przybylla, Mareen 49
Grillenberger, Andreas 207
Rahimi, Ebrahim 155
Heintz, Fredrik 117
Regnell, Björn 117
Hellas, Arto 15
Romeike, Ralf 27, 49, 207
Henning, Finn 49
Henze, Ineke 155
Hromkovič, Juraj 3, 219 Schreiber, Carla 49
Serafini, Giovanni 219
Jašková, Ľudmila 37 Solitro, Ugo 169
Jasute, Egle 141 Standl, Bernhard 180
Stankovičová, Mária 37
Staub, Jacqueline 219
Kastl, Petra 27
Stupuriene, Gabriele 232
Krisch, Oliver 27
Suhre, Cor 65
Lacher, Regula 3
Leinonen, Antti 15 van Veen, Klaas 65
Leinonen, Juho 15 Veselovská, Michaela 91
Lonati, Violetta 102 Vinikiene, Lina 232
Malchiodi, Dario 102 Zakauskas, Rimantas 232

Mannila, Linda 117 Zwaneveld, Bert 65

2017 Book InformaticsInSchoolsFocusOnLea

Uploaded by

Copyright:

Available Formats

2017 Book InformaticsInSchoolsFocusOnLea

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

2017 Book InformaticsInSchoolsFocusOnLea

Uploaded by

Copyright:

Available Formats

Valentina Dagiene

Arto Hellas (Eds.)

ISSN 0302-9743 ISSN 1611-3349 (electronic)

Library of Congress Control Number: 2017960853

LNCS Sublibrary: SL1 – Theoretical Computer Science and General Issues

© Springer International Publishing AG 2017

Printed on acid-free paper

This Springer imprint is published by Springer Nature

The International Conference on Informatics in Schools: Situation, Evolution and

October 2017 Valentina Dagienė

Conference Chairs and Co-chairs

Local Organizing Committee

The Computer Science Way of Thinking in Human History

Accessibility, Visualizations and Physical Computing

Adolescent and Adult Student Attitudes Towards Progress Visualizations. . . . 15

3D Printing as Medium for Motivation and Creativity in Computer

Teaching Materials in Informatics for Lower Secondary

Teachers’ Expectations and Experience in Physical Computing . . . . . . . . . . . 49

Teacher Development and Assessment

Investigating Informatics Teachers’ Initial Pedagogical Content

A Notation for Sets, Sequences and Series: Possible Benefits

Bebras Task Analysis in Category Little Beavers in Slovakia . . . . . . . . . . . . 91

Promoting Computational Thinking Skills: Would You Use

Introducing Programming and Digital Competence

Preparedness of Japan’s Elementary School Teachers for the Introduction

Baltic and Nordic K-12 Teacher Perspectives on Computational

Computational Thinking/Programming, Computational Abilities

Identifying Students’ Misconceptions on Basic Algorithmic Concepts

A Preliminary Investigation on Computational Abilities

Solving Everyday Challenges in a Computational Way of Thinking . . . . . . . 180

Teaching Basic Elements of OOP in School Informatics During

Real-Time Data Analyses in Secondary Schools Using a Block-Based

XLogoOnline: A Single-Page, Browser-Based Programming

Introduction to Bebras Challenge Management: Overview

Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245

Juraj Hromkovič(B) and Regula Lacher

Department of Computer Science, ETH Zurich, Zurich, Switzerland

Abstract. Teaching computer science oﬀers more than algorithmic

We consider this principle to be the base of the so-called “critical thinking,”

2.2 Computer Science and Mathematics

If one wants to understand what mathematics is about, it is helpful to view

The demand to create a language such as mathematics was obvious. There is no

There are so many intersections between mathematics and computer science

2.3 When and Why Computer Science Became an Independent

3 How to Teach Computer Science

(i) better understanding mathematics and languages and their development,

3.1 Data Representation

3.2 Algorithm Design

The main focus is on the following contributions:

Onni Aarne(B) , Petrus Peltola, Antti Leinonen, Juho Leinonen,

University of Helsinki, Helsinki, Finland

Abstract. Keeping students motivated for the duration of a course

Keywords: Progress visualization · Student engagement · Growth

2.1 Students’ Mindsets as Factors of Success

2.2 Point and Progress Visualizations

Fig. 1. Examples of the Weekly visualization, where the percentage of completed

Fig. 2. Examples of the Growth visualization where aggregate assignment score is

3.4 Research Questions

We have the following two research questions for this work:

– RQ1. Does the type of visualization aﬀect student scores?